CRELAND ON RIBS AND TRANSVERSE PROCESSES. 
123 
Before making any attempt to clear a subject, on wbicb so many 
theories have been formed, we shall do well distinctly to appreciate, 
that such discussions as that which engages our attention, as indeed 
all morphological questions, resolve themselves into investigations of 
the relative amount of significance to be attached to different classes 
of phenomena. We compare structures, and inquire in what respects 
they differ and in what they correspond. The question then arises ; 
what points of difference or correspondence are we to consider of 
primary importance, and what points are subordinate P The im¬ 
portance of such points can only be estimated by their prevalence 
in a series of animals, and the time of their appearance in the embryo. 
Thus, for example, we have already seen that the method of ossification 
and the exact place of origin of a part of a vertebra are not matters 
of primary importance, because structures closely related and com¬ 
parable, are found very variable in this respect. The point which I 
proceed to prove is that the distinction of greatest importance, in 
classifying the elements of a sclerotome, is that which divides parts 
into those which embrace the visceral cavity and those which do not. 
I start from this point, because all the theories mentioned above 
take for granted that the vertebral column is an axis to which typi¬ 
cally the neural arch is related on one aspect, as the visceral arch is 
related on the opposite aspect: a proposition totally at variance with 
that now* put forward. In fact, none of those theories can be held 
without considering the chorda dorsalis and bodies of the vertebrae as 
the pivot round which the symmetry of the whole animal is arranged. 
Now the only part in which the bodies of the vertebrae ever occupy 
such a position is the tail; and in it, especially in fishes, there is, no 
doubt, often an almost perfect supero-inferior symmetry pervading not 
the skeleton merely but likewise other systems. August Muller 
goes the length of saying “ Cauda ansa est morphologica.” But a 
segment of the tail cannot be taken as a typical segment of the body. 
We must take a part where all the layers of the embryo are repre¬ 
sented. The symmetrical arrangement of parts round the vertebral 
centra in the tail only shows that when the skeleton and muscular 
system are prolonged backwards beyond the viscera, the parts so 
prolonged tend to arrange themselves symmetrically round the bodies 
of the vertebrae. In the tail, not only is the intestinal canal absent, 
but the caudal vessels are no fair representative of the vascular 
system, which, in its typical disposition, completes circles round the 
digestive tube. Thus the aorta is not the whole morphological axis 
of the arterial system; the vessel which forms the heart is a prior 
development, and branchial arches are found at an early period in all 
vertebrata. Indeed the vascular system affords a key to the true 
fundamental arrangement of the different systems in relation to one 
another; viz. that they circle round the alimentary tube. This 
arrangement was recognised by Carus years ago, but mixed up, as 
his statements on this subject have been, with a method which led 
