252 REPORT—1846. 
their supporting skeleton in the primitive histological Fig 16, 
fibrous state, the corresponding parts are ossified in fishes: 
rarely, however, are such parts in answerable number to fg 
the vertebre; and the true spines of these vertebrae, a z 
when the median fins and their bony spines are removed, Sf) 8 
in fishes, show as little indication of the place or existence & E 
of such fins, as do the vertebre in the porpoise of the a 
existence of its dermal fin. In proportion as ossification 
has extended into the dermal system of fishes it has been 
arrested in the vertebre, which in the trunk and tail of 
fishes present their least complex condition. Two of the * 4 
autogenous elements, the ‘ hzmapophyses,’ are absent, and g 5 
are commonly represented, in the tail, by the modified % 5 
‘ parapophyses.’ ‘The seeming complexity of a fish’s ver- a = 
tebra arises from the intercalation of bones appertaining 
to the system of the dermo-skeleton : it would have been an 
unusual exception to the general course of development if 
the lowest of the vertebrate classes should have presented 
the vertebral skeleton in its highest state of complication ; 
and Geoffroy St. Hilaire was unfortunate in taking a fish’s 
vertebra with its extrinsic evertebrate complications, as the Po 
perfect type of that primary segment of the myelencepha- y: 
lous skeleton (fig.16). He was still more unlucky in having 
for the subject of his figure* a specimen from which two 
of the pieces, had been accidentally lost, as Cuvier after- 
Métapérial. 
Neural spine 
‘. Neurapophysis. 
- Cyclopérial. 
Cycléal. 
Centrum. 
wards pointed out ; yet Geoffroy’s mutilated caudal ver- . 2, |) s de 
tebra of the plaice continues to be copied in some 2 eae 
compilations of comparative anatomy, as the type ofa = & a 8 
vertebra! To obtain the dermal spines (pro-epial and pro- oS 5 é 
cataal) of the vertically extended caudal vertebre of fishes, a 
Geoffroy had recourse to a hypothetical division length- 
wise of the interneural and interhemal spines (which are - 
represented as being single in his figure), and to as gra- q | 3 
tuitous a displacement of one of the halves froin the side 8 \ 2 
to the summit of the other t. Now the interneural and & 2 
interhamal spines are actually double in relation to the om 
neural and hzmal spines ; yet they coexist with a dermo- 
neural and dermohzmal ray, which therefore needs no 
imaginary change of place of either of its supporting : 
spines to account for its existence. I subjoin in fig. g 3 
16 an entire vertebra answering to the mutilated one g 8 
figured by Geoffroy ; and for the better understanding of g e 
the difference between his determinations of the vertebral 8 
elements and those given in the present Report, the names 
respectively indicating those different determinations are 
added to the figure. In the description of the plate in Endo: and exo-ske- 
the ‘ Mémoires du Muséum,’ Geoffroy explains that the caudal, vertelsa of) 
‘ pro-épial’ is the left half or ‘épial gauche,’ and the en-épial 9, 7aice (Pleuro- 
the right half or‘ épial droit’ : that the en-cataal is the right 
half or ‘ cataal droit,’ and the pro-cataal the left half or ‘ cataal gauche,’ of his 
imaginarily divided epivertebral and catavertebral elements (i. ep. 115) 
* Mémoires du Muséum, t. ix. (1822), pl. 5, fig. 1. 
+“ Lune de ces pieces monte sur l’autre”—‘ l'une se maintient en dedans, quand 
l’autre s’élance en dehors,” id. p. 97. 
