CLELAND ON RIBS AND TRANSVERSE PROCESSES. 
119 
position; but in certain cases they present only the former articula¬ 
tion, and in other cases they are described, not without error, as 
having only the latter. 
On the other hand, in fishes those transverse processes which bear 
ribs tending to surround the visceral cavity are unconnected with the 
neural arch. They are in serial continuation with the inferior arches 
of the caudal vertebrae; and the ribs themselves are also in continua¬ 
tion with the distal portions of these arches or with the spines into 
which they are prolonged. But while this is the disposition of the 
transverse processes and ribs embracing the abdominal cavity, certain 
fishes present processes in the tail, which, lying in the lateral inter¬ 
muscular septum, strike out transversely like the caudal transverse 
processes of mammals. Also in the same intermuscular septum there 
are in certain fishes articulated rib-like bones which strike towards 
the skin, and are attached directly to the vertebrae, or, in some cases 
in the forepart of the trunk, to the upper aspects of the ordinary 
ribs. They are sometimes described as the superior range of ribs. 
Besides all these, there are, in many fishes, rows of bones generally 
admitted to be of only secondary importance, disposed one row above 
and another row below the middle lateral line. They are very fully 
developed in the herring, in the trunk of which the superior row are 
given off from the bases of the neural arches, and the inferior row 
are attached by ligament to the bodies of the vertebrae, while both 
rows are continuous at the commencement of the tail with bones 
more loosely connected with the vertebrae, and forked at their proxi¬ 
mal extremities, and these are continuous at the back of the tail with 
others in which the forked condition has disappeared, and which lie 
almost entirely on the surface of the muscular mass. 
Johannes Muller, # to explain the differences of arrangement of 
ribs and transverse processes in fishes and other vertebrata, dis¬ 
tinguished no less than four kinds of transverse processes: viz. the 
ordinary mammalian, attached to the neural arch; the ordinary 
piscine, forming the haemal arch, to which he also referred the 
inferior arches in tails of mammals and reptiles; a row superior to 
the ordinary mammalian, viz. the mammillaries ; and a row superior 
to the ordinary piscine, to which belong only the superior transverse 
processes of fishes, and the lumbar and caudal transverse processes of 
the cetacea, which differ from those of other mammals in being at¬ 
tached to the centra without connexion with the neural arches.f He 
maintained that it was characteristic of a rib to embrace the visceral 
cavity, and therefore considered that the piscine and mammalian ribs 
* Vergleichende Anatomie der Myxinoiden, p. 100. 
f Muller indeed states (Op. cit. p. 100), “ the transverse process of the lumbar 
and caudal vertebra in the cetacea is developed merely from the centrum of the ver¬ 
tebra, as I have satisfied myself in the young Narwhal:” but in a foetal cetacean of 
uncertain species I find that it is ossified from a centre of its own. 
