DR. A. GUNTHER’S DESCRIPTION OF CERATODUS. 
529 
in its hind part. Like the neural and hsemal spines, and other parts of the skeleton, 
the basal and distal portions of the rib remain cartilaginous whilst the central part is 
enclosed in bone. The terminal cartilages penetrate for some distance into this bone, 
but without being continuous, the centre of the bone being filled with cellular tissue 
and marrow. There is no real joint between the rib and hsemapophysis ; its cartilagi- 
nous head fits into a shallow concavity of the latter, both surfaces being united by short 
connective tissue, allowing no free motion to the rib. The first rib (Plate XXX. fig. 2, 
Plate XXXIV. fig. 3, x, and Plate XXXVIII. fig. 3) deserves to be particularly noticed, 
not only because it somewhat differs in shape from the others, but because the corre- 
sponding bone in Lepidosiren has received various interpretations. It is a long bone, 
considerably thicker and more cylindrical than the other ribs, horizontally directed out- 
wards and slightly backwards, forming the posterior lower limit of the gill-cavity. Its 
insertion is opposite to the first neurapophysis, being joined to the first hsemapophysis, 
and not to the basal bone, which at this place is slightly contracted. It crosses the 
suprascapula, from which a short ligament passes to a small cartilaginous protube- 
rance (V) on the front edge of the rib. This tubercle, which is placed exactly at the 
spot where those two bones cross each other, has evidently only a functional, and not a 
homological significance. Peters (Mull. Arch. 1845, p. 12, pi. 2. no. p) is the first who 
distinctly notices this bone in Protopterus , as “ a peculiar bone which is to be com- 
pared to a similar bone in Patrachus .” Bischoff (Ann. Sc. Nat. 1840, vol. xiii. 
pp. 123, 126) represents the neurapophyses of the first vertebral segment in Lepidosiren 
paradoxa as occipitalia lateralia, and the neural spine of the same segment as a cartilage 
“ qui remplace, en quelque sorte, l’ecaille de 1’os occipital consequently he describes 
the first rib as inserted in the occipital lateral and in the body of the sphenoid, declaring 
it to be the “ os suspenseur de la ceinture pectorale des poissons.” According to this view, 
suprascapulary elements would be absent in Lepidosiren paradoxa ; future investigations 
must show whether this is really the case. Hyrtl and Owen (Anat. Vert. i. p. 83, 
fig. 41. no. 51) have adopted Bischoff’s view; and the bone is figured and described by 
the latter as scapula, in direct connexion with the pectoral arch. Finally, W. K. Parker 
(‘ Monograph of the Shoulder-girdle,’ 1868, p. 21) not only determines the homology of 
this bone, but also of the small cartilage attached to the middle of its length ; the former 
is stated to represent the “ large first pharyngo-branchial,” the latter the “ small unossi- 
fied second pharyngo-branchial.” 
A comparison of Lepidosiren with Ceratodus shows that a positively defined boundary 
between the vertebral column and skull does not exist, that parts which in one genus 
appear to belong to the skull, are distinctly portions of the vertebral column in the other, 
and that, with regard to the particular bone which has been so differently interpreted 
by the authors named, the opinion first expressed was nearest to the truth. The peculiar 
bone in Patrachus, to which Peters directed attention, is in fact nothing but the first 
rib (see Gunther, ‘Fishes,’ vol. iii. pp. 167, 172), extending and fixed to the upper end 
of the humerus. Although its insertion into the top of the neurapophysis is a most 
4 d 2 
