126 Royal Society : — 



palate. Is about 6 inches ; the length of the mandible can hardly have 

 been less than 1 2 inches. 



Of the vertebral column, the greater part of the sacral and dorsal 

 region, and some fragments of the cervical region, are preserved. The 

 latter show that the atlas was distinct, but that the axis was anchy- 

 losed with one or two succeeding vertebrae, as in the Armadillos. 

 The fifth and sixth cervical vertebrae were probably free, but no traces 

 of them have been found. The anterior part of what remains of 

 the rest of the vertebral column consists of a very broad flat bone, 

 composed of three vertebrae firmly anchylosed together, and having 

 their spinous processes represented by a short but very stout osseous 

 knob, which projects upwards and backwards. Anteriorly, these 

 anchylosed vertebrao exhibit on each side of the neural canal an arti- 

 cular facet with a convex surface, resembling a segment of a horizontal 

 cylinder ; posteriorly, articular surfaces of a similar character, but 

 concave, are situated in corresponding positions. 



Each side of this ' trivertebral bone' presents two large and deep 

 articular cavities for the heads of ribs, fragments of which are still 

 preserved. The anterior rib, remarkable for its stout and massive 

 proportions, was undoubtedly the first ; and this circumstance I believe 

 gives a clue to the precise character of the vertebrae which are anchy- 

 losed together to form the trivertebral bone ; for in the Armadillos 

 the head of the first rib is fitted into a deep fossa, formed partly by 

 the last cervical, and partly by the first dorsal vertebra. Furthermore, 

 the body and transverse processes of the last cervical vertebra in the 

 Armadillos present articular facets of an essentially similar character to 

 those observable on the anterior face of the bone under description* ; 

 and, finally, the last cervical vertebra is practically immoveable upon 

 the first dorsal in many Armadillos, while the two vertebrae are com- 

 pletely anchylosed together in the priodont Armadillo. I conceive, 

 then, that this remarkable bone of the Glyptodon is formed by the 

 anchylosis of the last cervical and first and second dorsal vertebrae. 



Of the remainder of the spinal column thirteen consecutive ver- 

 tebrae are preserved ; and all of these were immoveablv united into 

 one long continuous tunnel or arched tubular bridge of bone, a struc- 

 ture which is without a parallel among the Mammalian Vertebrata. 

 Of these thirteen vertebrae, the four .anterior are so completely an- 

 chylosed together, that the original lines of demarcation between 

 them are hardly discernible. Persistent sutures separate the fourth 

 from the fifth, and the latter from the sixth ; but all trace of the 

 primitive distinction of the sixth and seventh is lost. The other 

 vertebrae are separated by sutures which become coarser and less close 

 posteriorly. In all but the first, second, third, eleventh, and thir- 

 teenth vertebrae, the parts representing the vertebral centra are broken 

 away ; but where they persist, they are so similar that they were 

 doubtless of similar form throughout. Each centrum is, in fact, a 



* I may remark in passing, that all the cervical vertebrae of the Armadillos, 

 from the third backwards, are articulated together by joints similar in principle 

 of construction to those which connected together the trivertebral bone of Glyp' 

 todon with the vertebrae in front of and behind it. 



