128 Royal Society : — 



The nervous foramina are not intervertebral, but pierce the arches 

 of the vertebrse throughout the series. In the thirteenth the outlet 

 of the foramen is separated, by a longitudinal bar of bone, into an 

 upper and a lower division. 



The posterior part of the thirteenth vertebra is much injured, and 

 does not adjust itself naturally to the anterior end of that part of the 

 lumbar region of the vertebral column (consisting of two vertebrse) 

 which remains continuously anchylosed with the sacrum. One or 

 two vertebrae may possibly be wanting, or even three ; but I conceive 

 the last to be the extreme limit of the deficiency*. 



The great Priodont Armadillo has twenty dorso-lumbar vertebrae. 

 If the Ghjptodon had the same number, there would be three missing ; 

 for there are two dorsal vertebrse in the trivertebral plate, thirteen 

 follow it, and two lumbar are anchylosed with the sacral, making 

 altogether seventeen. 



The ' sacrum,' composed of anchylosed lumbar, proper sacral, and 

 coccygeal vertebrse, contains at fewest twelve, and perhaps thirteen 

 vertebrse. The centra of the two lumbar vertebrse and of the two 

 proper sacral vertebrse which follow them are preserved. They are 

 thin and broad plates, flat above and slightly concave below, exhibit- 

 ing a most marked contrast with the half-cylinder of the hindermost 

 of the thirteen dorsal vertebrse above described. It would seem to 

 require the interposition of at least two, if not three, vertebrae to 

 eflFect the transition of the one form of centrum into the other. 



The last coccygeal is the only vertebra among all those preserved 

 the centrum of which exhibits characters at all like those of an or- 

 dinary mammal, its terminal face being a very broad oval, slightly 

 concave, disk. The centrum of the penultimate coccygeal is n)uch 

 flatter and narrower ; and this flattening and narrowing predominates 

 still more in the antepenultimate and that vertebra which lies before 

 it, or the fourth from the end. From this point to the two anterior 

 sacrals the floor of the vertebral canal is completely broken away, but 

 there can be no doubt that the centra were represented by a thin 

 bony plate. 



The line of the centra of the coccygeal vertebrse forms a very 

 marked arch behind the two sacral vertebrse, whose centra form a 

 nearly horizontal floor ; while the dorso-lumbar vertebrae (including 

 the trivertebral bone) form a second arch, flatter than the first. 



The spinous processes of all these lumbo-sacro-coccygeal vertebrse, 

 up to the fourth from the end inclusively, are anchylosed together 

 in a long and strong osseoiis crest, broad and extremely rugose above, 

 eight inches high in front, but slowly diminishing as it follows the 

 curve of the centra posteriorly to five inches. 



The spinous process of the penultimate coccygeal vertebra is very 

 thick, but is broken short off. It was probably not less than 4 inches 

 high, and afl"orded a middle point of support for the carapace between 

 the ischial protuberances. The sides of the median crest, and of the 

 two vertebrae which appear to constitute the true sacrum, are anchy- 



* Unless I greatly err in my interpretation of the photographs, these three 

 missing vertebrse are preserved in the Turin Glt/jitodon. 



