Dr. R. H. Traquair—Homosteus and Coccosteus compared. 5 
space open anteriorly, and in this space is lodged, first a small 
elliptical plate, the posterior ethmoidal (p. e.) and the median limb 
or stalk of the anterior ethmoidal (a. e.). This latter bone, the “ pre- 
maxilla” of Huxley, is like a nail-head with a broken off stalk 
attached, the head forming the anterior point of the buckler, the 
stalk passing back between the pre-orbitals. On each side of it in 
front are the small nasal openings (n.), each being bounded externally 
by a small separate bone, the premawilla (p. ma.). The orbit is 
bounded below by the “paddle-shaped” bone (ma. Fig. 3) repre- 
senting the mazilla, which has a sort of resemblance to that in 
Palgoniscus, and to the posterior extremity of which is appended a 
small triangular plate (j.), the jugal or post-maxillary. 
Turning now to Homosteus, we shall find that Hugh Miller’s 
comparison of its cranial plates with those of Coccosteus is not so 
far amiss, of course, taking into account the faults of his reading of 
the buckler of the latter genus. In Fig. 1 I have sketched the 
configuration of the specimen of Homosteus, referred to by John 
Miller as having the dorsal plate in situ. Much of the bony sub- 
stance having splintered off from the “specimen ” itself, I have had 
a plaster impression taken from the ‘“ counterpart,” which con- 
sequently reproduces the details of the original in a very much 
more perfect manner, and from this model the drawing has been 
taken. And I may add that every detail of the buckler here given is 
corroborated by another splendid specimen, also from the collection 
of John Miller, in which, however, the dorsal plate has got dis- 
placed to one side. 
We find a wonderful correspondence in the arrangement of the 
bony plates, the differences appearing almost entirely due to the 
altered position of the eyes, and the assumption by the cranial shield 
of an antero-posteriorly elongated, instead of a broadly hexagonal 
figure. 
The median occipital (m. o.), preserving its trapezoidal shape, has 
become much elongated, as have also the external occipitals (e. 0.), 
while the centrals (¢.) have become much smaller in proportion, and 
have come to take part in the inner boundary of the orbit, more, 
however. on the upper than on the under aspect; they are also in 
contact in front with the posterior ethmoidal (pt. e.), the hinder 
angle of which is inserted in a notch between their anterior ex- 
tremities. The marginal (m.), also much elongated, is easily 
recognized ; but it is now alongside of, instead of in front of, the 
external occipital, and the post-orbital (pt. 0.) and pre-orbital (p. 0.) 
have altered their relation to the orbit in a strange fashion. 
Separating internally, so as to allow the central to come into the 
boundary of that opening, they have thrown out processes which 
unite externally, and so in Homosteus the orbits come to be entirely 
enclosed within the buckler, instead of being outside it, as in 
Coccosteus. The last plate to be noticed is the anterior ethmoidal 
(a. e.), which occupies a position at the front of the snout exactly 
as in Coccosteus, but no trace is seen either of the small pre- 
maxillary, or of the nasal openings of that genus. 
