100 
tractions produce sometimes additional folds and elevations. (PL. 1v. fig. 1. 
toll.) Ifany doubt should still be entertained whether these folds and mark- 
ings on the plates are really derived from muscular contraction, on the ground 
that they may have possibly resulted from the original configuration of the solid 
calcareous matter which forms them ; a general view of their phenomena, and 
change of figure under various circumstances, in a manner which may be readily 
accounted for on the hypothesis of their muscularity, but not on any other, 
will, I am persuaded, remove these doubts; but I have it in my power to 
adduce a proof still more decisive. There are some costal plates in my pos- 
session which appear to have been torn from the animal with violence while 
living, or before decomposition took place. The consequence of this would 
naturally be that the lacerated muscle would shrink from the edges of the 
plates towards its point of adhesion in the centre ; and accordingly this is 
exactly the appearance preserved in the fossil specimens ; the surface where 
denuded of its muscle, being irregularly corrugated, (Pt. 111. fig. 15. 16. and 
20.) whilst the muscular folds are gathered into the centre. 
As the pelvis supports six plates, it might have been reasonably concluded 
that it was the intention of nature to sustain six arms; and as the animal has 
actually only five, that the omission must either create a great vacancy, 
or give rise to a necessary alteration of the general mechanism, in order to 
render the circular net, formed by the arms and fingers when extended, com- 
plete. Nature however is never at a loss, but accomplishes her purposes with 
ease, notwithstanding the new difficulties that are created by continually vary- 
ing structures. Thus in this animal the apparent difficulty so presented is 
obviated by the change of form in the supernumerary costal which is penta- 
gonal, while the other five are hexagonal. From the general arrangement of 
the plates, these five hexagonal costals can only give rise to a series of second 
costals, terminating in scapulz and arms, (PL. 11. fig. 1.) while the irregular or 
sixth costal plate intervening between them, (PL. 11. fig.2.) having a pentagonal 
form, and presenting only the edges of its upper angle to the next row, supports 
two series of intercostal plates occupying the interval, occasioned by its interpo- 
lation, and so formed that this interval in the succeeding rows gradually 
diminishes in proportion to the whole circumference, so that although there 
is still a somewhat greater distance here between the two scapule placed on 
either side of it, than between the other scapula, yet the irregularity is not such 
as to occasion any material inconvenience or interruption of symmetry. The 
