170 
MR, J. W. 1IULKE OX POLACAXTHES FOXII. 
is now evident that the piece's which I described in my paper of 1881 were not all 
integers of the value of separate scutes, entering by overlap, or other arthrodial 
articulation, into the composition of the shield, but (with the exception of the y spines) 
were pieces originally synostosed in a continuous sheet, in which no traces of suture 
or other marks of primitive distinctness are discernible. 
Near its lateral and posterior border the shield is thicker than at the middle, where 
it overlies the vertebral column, a fact noticed by Mr. Fox. The form of these 
borders is that of a smoothly-rounded lip separated from the upper surface by a 
narrow sunken groove. The continuous sweep of the posterior border is interrupted by 
a wide deep notch, having at its centre a projecting part that overhung the root of 
the tail. At each side of this projection is a narrow cleft ; whether natural, or a 
crack produced by the yielding of the shield under compression, is uncertain. The 
anterior differs from the other borders of the shield by its attenuation to a thin edge. 
This circumstance, together with the truncated form of the border, suggests that the 
part we actually possess does not represent the complete dorsal mail, but only a 
posterior segment, between which and an anterior part, now lost, covering the thorax, 
a flexible junction may have existed somewhat like that present in the plastron of 
certain Chelonians. 
The upper, or exterior, surface of the shield is richly ornamented. It exhibits in a 
highly satisfactory manner the position and distribution of the pieces described by me 
in 1881 as “ tuberculated ” (a) and “ keeled ” (fi) scutes. The former (a) compose 
the general groundwork co-extensive with the whole area of the shield, which is 
closely studded with hemispherical tubercles averaging 5 to 1*5 centims. across their 
base. The /3 pieces, characterised by a keel-shaped elevation rising out of a circular 
or elliptic depression, are grouped in four longitudinal rows occupying each lateral half 
of the shield. The highest and stoutest part of the keel is always posterior. The 
largest keels form a sub-marginal lateral row ; and the smallest a paired sub-median 
series, one in each pair lying on each side of the middle line of the shield over the 
vertebral column. The central elevation in this latter series resembles a low blunt 
cone, with circular or oval base, rather than a keel. The keels of intermediate size 
compose two less regular rows placed between the sub-marginal, lateral, and the sub¬ 
median series. It is now certain that the large spines (y) do not constitute any part 
of the shield we possess ; and since, for reasons given in 1881, they appear excluded 
from the caudal mail, they would seem to have occupied an anterior region of the 
trunk, a supposition which derives some additional probability from the position of 
the series of similar spines in the type-specimen of ILjlceosaurus preserved in the 
British Museum. 
Thin sections of the shield, mounted in Canada balsam, show an osseous structure. 
Near the inner surface the arrangement of the trabeculae conforms to that of the 
decussating bundles of fibrous tissue observable in the cutis vera of existing Lizards. 
Near the outer surface sections, vertical to and parallel to this surface, show an 
