160 SHUFELDT. 



These anapophysial processes of the lumbar vertebra, where present, 

 are quite chara,cteristic. Each one on either side forms a deep notch 

 with the postzygapophysis next to it on the same vertebra. Into this 

 notch, when the bones of the spine are normally articulated, fits the 

 prezygapophysis (of that side) of the next succeeding vertebra, the com- 

 bination making a very close interlocking articulation, which, when taken 

 altogether for the six leading lumbars, accounts for the remarkable fixity 

 and stability of this part of the column in Cynocephalus. In the last 

 two lumbar vertebrte these anapophysial processes are entirely aborted, 

 as are the transverse processes in the first two lumbars. After that, 

 however, these diapophyses begin to appear again, being represented by 

 thin, quadrilateral, horizontal plates of bone of good proportions. They 

 are thick and strong in the last lumbar, and claw-shaped in the three that 

 precede it with the apices of the claws directed to the front. (Plate IV, 

 figure 13.) Metapophyses of a very rudimentary tjrpe are also to be seen 

 in the mid-lumbar vertebrse; in any case the most of the projection belongs 

 in reality to the anterior zygapophysis, an. exception being found in the 

 last lumbar vertebra, where these processes are much better defined and 

 rather more prominent. 



In Steere's specimen the sacnim is a very solid bone, composed of three 

 vertebrae thoroughly fused together. The leading one has double the bulk 

 of the last, while the middle one is massive anteriorly and slopes away 

 behind (Plate IV, figure 14). Anteriorly, the first sacral presents the 

 usual facets, processes, and surfaces to articulate with those on the hinder 

 aspect of the last lumbar. So, too, the posterior face of the third sacral 

 is similarly modified in order to meet the requirements of an articulation 

 with the anterior face of the first caudal vertebra. Laterally, the entire 

 mass of the first sacral and the anterior moiety of the second, are enlarged, 

 thickened, and curved ventrad to support on their outer aspect a large, 

 subelliptical, articular surface for the ilium of the same side. This 

 surface looks upward and outward, and the major axis of the ellipse is 

 in line with the longitudinal axis of the spinal column, being parallel to 

 the long axis of the ilium when the bones are normally articulated. There 

 are two pairs of foramina on the ventral surface of the sacrum for the 

 exit of the anterior roots of the spinal nerves. Similar pairs of foramina, 

 for the posterior roots, occur on the dorsal aspect of the bone directly 

 opposite these; then the foramina for the pairs of roots of the spinal 

 nerves both anterior and posterior to these are only completed when the 

 last lumbar and first caudal vertebrae are in position and duly articulated 

 in the column, being represented only by shallow notches in front and 

 behind when we study the sacrum as a single bone. 



Considerable compression in the vertical direction is noticed in the 

 first four caudal vertebrae. The fifth is rather stocky, after which they 

 commence to elongate and lose their apophyses and the other usual 



