494 PROCEEDINGS OE THE GEOLOGICAL SOCIETY. [June 13 



is a wide transverse ellipse, with outlets encroached on below by the 

 convex upper border of the articular ends of the centrum. This en- 

 croaching part of the centrum contracts as it extends into the canal, 

 and then again expands to the opposite end ; so that the floor of 

 the neural canal shows a median low rising of bone expanded at both 

 ends, the curved depressions on each side indicating the original 

 separate bases of the neurapophyses. 



I subsequently received from Mr. Moore a few fossils from the 

 mixed deposits of the fissure above-mentioned, including the cen- 

 trum of a similar vertebra, rather smaller in size, with the median 

 part of the lower surface less ridge-like, the part convex across being 

 broader than in the preceding vertebra. The terminal articular cups 

 were deep. The bases of the neural arch span outwards, as far as they 

 are preserved, indicating the same expanse of the neural area as in 

 the more perfect vertebra above described. The floor of the canal 

 shows two lateral curved lines or slight depressions, with the con- 

 vexities turned towards each other, and which indicate the original 

 suture between the centrum and neurapophyses. 



In the same collection was the fore part or half of a centrum with 

 the anchylosed bases of the neurapophyses, showing a longitudinal 

 canal on the under part of the centrum, formed by two nearly pa- 

 rallel, longitudinal, low but sharp ridges, which slightly diverge 

 near the articular end. This presents a deep hemispheric cup, smooth, 

 with sharp margins. The transverse process begins at the lower or 

 haemal ridge, in the form of a narrow ridge, and ascends obliquely 

 forward to the side of the neural arch, where it becomes thicker ; but 

 the arch is broken away at this part, so that no articular surface is 

 preserved on the transverse process, if such existed. It most pro- 

 bably projected as a free process ; for the modification of the lower 

 surface would indicate an anterior caudal vertebra. 



With the above portion of vertebra was preserved a cupped arti- 

 cular end of a centrum, which may have belonged to the same ver- 

 tebra. 



Of the above three vertebras the second is certainly, and the third 

 most probably, of the same species as the first and most perfect spe- 

 cimen, above figured. 



No known Mammal, recent or extinct, has hitherto presented the 

 biconcave structure shown by these little vertebrae from the ossiferous 

 fissure near Frome. But such structure is known, as a very rare 

 exception, amongst existing Saurians ; and it might be asked, since 

 the biconcave structure was the ride among Secondary Saurians, why 

 it may not have prevailed with the Secondary Mammals ? ■ 



Hitherto we know nothing of the vertebral characters of the rare 

 Oolitic and Triassic Mammalia ; and unless the problematical ver- 

 tebrae above described do belong to the little Mammal with whose 

 teeth they are associated, we have still to learn the vertebral cha- 

 racters of Microlestes, Ampliitlier'mm,Spcdacotliermm, Triconoclon, &c. 



The only approach (and it is a slight one) made by Mammals to 

 any of the Saurian characters in the vertebrae above described is, 

 so far as I know, to be met with in the Monotremes, which offer in 



