THE OLD RED SANDSTONE FISHES OF ENGLAND. 443 
detected in any specimens, and palaeontologists therefore con- 
clude that the month was soft and probably used as a sucker, 
being placed on the under surface of the head, like that of the 
sturgeon. The remarkable position in which the eyes are 
placed, would, however, seem to render them useless as regards 
the detection of food if this were the case. The body which 
was attached to this head has almost as remarkable an appear- 
ance as the head itself ; that is to say, it is remarkable when 
viewed by the side of those soft-bodied, degenerate fish which 
now inhabit our waters. A great variety of species of fossil 
fish possessed bony scales, and the Cephalaspis was one of 
these. The body was rather small compared to the size 
of the head, and tapering towards the tail (fig. 3). The 
scales were arranged in three sets on each side, those on 
the flank being long plates ; they were all so arranged that 
the fish had the appearance of being enclosed in a suit of 
mail, “ cuirassee," as the great icthyologist Agassiz termed 
it when describing the first known specimen. The markings 
on the surface of the scales are not well known, since the stone 
is so hard that they are destroyed in fracturing the rock. Of 
the fins of the Cephalaspis we only know two. These are 
the caudal fin or tail, and the dorsal. The caudal fin is 
supported by several rays, which spring from the lower side 
of the tapering, pointed termination of the body. It is of that 
kind known as heterocercal, such as the sharks and dog-fish 
possess now-a-days ; whilst most recent fish have a homocercal 
tail, in which the fin rays instead of springing from the lower 
side of the vertebral column, arise from it equally above and 
below. All very young fish at first have a heterocercal tail, 
which afterwards in most recent forms becomes homocercal ; 
and it is somewhat interesting to find this peculiarity of 
the early stage of a fish's life, characterizing the earliest 
fish which are known to us. The dorsal fin was placed 
near the tail, and was rather long. In two specimens 
which we have seen, there are traces of a membrane con- 
necting the two lateral horns of the head-disc with the 
body. These may have represented pectoral fins, but better 
specimens will one day, it is to be hoped, show us more of 
them. The microscopical structure of the shield of Cephal- 
aspis is interesting, since it affords one great reason for 
separating this fish from Pteraspis. The bone of which the 
head is composed was covered by a membrane or skin of some 
description which was very full of blood-vessels, since we can 
see the canals in which they ramified on the surface of the 
shield. In some specimens we find an appearance like a 
number of hexagonal plates, or minute starlike bodies (fig. 8) 
on the specimens of the Cephalaspid's head. These were at 
