258 REPORT—1846. 
are formed by the gradually bent-down ribs*, which are formed in the 
abdomen either by unusally elongated ‘parapophyses’ (if they be inter- 
preted by the condition of those elements in the cod-fish), or by pleurapo- 
physes articulated directly to the fibrous sheath of the notochord; which 
interpretation of the mode of formation of the hemalarches is supported by 
Professor Miiller’s discovery of the nature of those arches in the Lepidosteus +. 
Whether we adopt the analogy of the Anacanthini, or the Ganoidei (and 
the general affinity of the Protopteri to the ganoids would incline the choice 
to the latter), the constitution of the hzmal arches in the lepidosiren is 
strictly piscine; at least if we take the skeleton of the tailed batrachia 
(tig. 28) as our guide to the homology of the caudal inferior arches in 
higher reptiles and mammals. The unusual size and length of the abdo- 
minal parapophyses in the cod-tribe ( Gadide), the flat-fishes (Pleuronectide), 
and the genus Ophidium, evinces the natural character of the order Anacan- 
thini, in which they have been grouped together by Professor Miller: the 
pleurapophyses are, conversely, very short and slender in this order. In all 
bony fishes the costal arch in the abdomen is completed by the aponeurotic 
septa between the ventral portions of the myocominatat, which there repre- 
sent the ‘ hemapophyses’ (cartilagines coste, inscriptiones tendinee muse. recti 
abdominis of anthropotomy). Indeed, when we reflect that the trunk of 
the fish, by reason of the advanced position of the heart and breathing organs, 
answers to the abdominal and caudal regions of the trunk of higher verte- 
brates, we could hardly expect the typical vertebra to be there carried out in 
osseous tissue; but rather be prepared to find the hemapophyses retaining 
the same primitive histological state which they present in the abdomen of 
mammals and man (fig. 25, h'’). ; 
Immediately behind the coracoid arch, it is usual to find a long and slender 
rib-like bone, sometimes composed of two pieces, on each side; it gives a 
firmer implantation to the portion of the myocommata immediately behind 
the pectoral fin; and is obviously the ossified serial homologue of the hema- 
pophysial aponeuroses between the succeeding myocommata. It is usually 
detached from its centrum and articulated superiorly to the inner side of the 
coracoid: when it rises higher, as in the Batrachus, it becomes attached to 
the atlas, and in the Argyreiosus vomer it meets and joins its fellow below, 
forming a true inverted or hemal arch, parallel with, but more slender than 
the coracoid arch. No other idea of the general homology of this arch pre- 
sents itself than as a hemal one, completing the costal arch as an ossified 
hemapophysis, differing from the typical vertebra (fig. 15) only by the non- 
development of a sternum or hemal spine: and there appears to be as little 
ground for hesitation as to the particular segment of the endoskeleton to which 
to refer this costal or inverted arch; its immediate succession to the correspond- 
ing arch attached to the occiput, as well as the occasional direct attachment, 
indicating that segment to be the atlas or first vertebra of the trunk. 
The best-marked general character of the vertebral column of the trunk in 
the class Pisces is that which Professor J. Miiller first pointed out ; viz. the 
formation of the hemal arches in the tail by the gradual bending down and 
coalescence of the parapophyses ; the exceptions being offered by the ganoid 
polypterus and lepidosteus and the protopterous lepidosiren. The pleurapo- 
physes are, sometimes, continued in ordinary osseous fishes from the parapo- 
physes after the transmutation of these into the hemal arches. The dory, 
* Linn. Trans. vol. xviii. pl. 23, fig. 4, 2 2. 
. + Remarks on the Structure of the Ganoidei, in Taylor’s Scientific Memoirs, vol. iv. 
p. 551. 
+ Lectures on Vertebrata, 1846, p. 163, fig. 44, h p. 
