ON THE VERTEBRATE SKELETON. 281 



the haemal spine (22) to the nasal centrum (13), and by the upward develop- 

 ment of the processes of 22 which join the neural spine (is). Much modifi- 

 cation was to be expected in the segment which terminates the skeleton 

 anteriorly ; and yet the typical characters of the neural arch are more com- 

 pletely preserved here than at the opposite end of the vertebral column. If 

 the bones 4, s, 12, which I recognise as ' parapophyses ' in the cranial 

 segments 1, 11, in, must be viewed as superadded intercalations for the 

 special and characteristic expansion of the neural arches of those segments — 

 normal elements, indeed, of the typical vertebra, but withmodified connections 

 for cranial functions — then the disappearance of their homotypes in the nasal 

 segment restores its neural arch (fig. 4) to the more common condition, and we 

 recognise in 13 the centrum, in 14, 14, the new -apophyses, and in 15 the neural 

 spine of the nasal vertebra. 



But the segment to be complete should exhibit a second arch, inverted ; and 

 we find such arch closed or completed by the symphysis of the bones 22, 

 fig, 5, and suspended to the sides of the centrum 13 and to the neurapophyses 

 14, 14, by the bones 20, as the piers or crura of the arch ; these bones being 

 connected to the key-bones 22, by the intermediate bones 21. Now, the 

 modifications which these elements of the inverted or haemal arch of the 

 nasal vertebra have undergone, are, also, much less than might have been 

 anticipated from the extent to which the segments are modified at the oppo- 

 site extreme of the endoskeleton. All the normal elements of the haemal 

 arch, for example, are retained : 20 is the pleur apophysis, 21 the h&mapo- 

 physis, and 22 the hcemal spine, in most fishes divided at the middle line, but 

 sometimes confluent with its fellow e. g. Diodon. The essential (pleur- 

 apophysial) part of 20 extends in many fishes (e. g. percoids) like a short 

 straight rib from its articulation with 13 and 14 to the condyle at its opposite 

 end to which the haemapophysis 20 is articulated ; but it usually, also, de- 

 velopes a process from its hinder margin downwards and backwards, which 

 gives attachment to the diverging appendage of the arch H iv. The de- 

 velopment of the other bones of the arch, 21 and 22, outwards, downwards 

 and backwards, is still more marked in relation to the protractile and retrac- 

 tile movements of the arch in most osseous fishes ; and some anatomists, 

 influenced by the form and proportions rather than the connections of those 

 bones, have described them as independent parallel arches : but, as such, 

 they must be regarded as being suspended by their apices or key-stones to 

 the axis of the skull, and as having their haunches hanging freely downwards 

 and outwards — a position the reverse of that of the foregoing inferior arches 

 of the skull and of every typical haemal arch. The reduction of that di- 

 vergent development, characteristic of the bones 21 and 22 in fishes, is ef- 

 fected in a great degree within the limits of the piscine class : already we 

 find one of the spurious arches abrogated in the salmonoid fishes by the short- 

 ening of 22, and its more direct continuation from 21, which now forms the 

 larger part of the upper border of the mouth and supports teeth : the con- 

 fluent maxillaries and premaxillaries send down only a single divergent 

 process from their point of suspension to the palatine condyle in the plecto- 

 gnathic fishes; and the consolidation of all the elements of the palato-maxillary 

 arch into its normal unity is effected in the lepidosiren*. The palatines (20) 

 always form the true bases or suspensory piers of the inverted haemal arch 

 at their points of attachment to the prefrontals (14) ; the premaxillaries, 22, 

 constitute the true apex or crown at their symphysis or point of confluence, 

 H iv ; the approximation of which to the anterior end of the axis of the skull 

 is rendered possible, in fishes, by the absence of any air-passage or nasal 



* Hunterian Lectures, Verlebrata, p, 81, fig. 29. 

 184-6. u 



