The metatarsus of the Solitaire differs from that of the Dodo, not only in the greater elongation and
antero-posterior expansion of the central portion of the shaft, but also in the greater breadth and transverse
flatness of the-external border, or surface uncovered by muscle, which does not curve round the upper part
of the tumid external metatarsal pillar, as in the Dodo, but. encroaches on and flattens that element, so that
this margin is concave vertically in the Solitaire; while in the Dodo, it is slightly convex in its
upper moiety; its anterior and posterior edges are acute, but the anterior in the Dodo is rounded off.
Theconcavity beneath the proximal extremity is deeper, and its floor angular ; the outer wall formed by the
central and external elements is only slightly concave; but the inner pillar is more convex and tumid than
•in the Dodo. The rounded surface for the attachment of the Tibialis anticus is in contact with the internal
inter-osseous foramen, and extends on both walls of the concavity, a deep groove bisecting it; the outer
segment is slightly raised, the inner impressed. The groove for the tendon of the M. adductor
annularis is shorter and less distinct; and the tendon is transmitted through a canal, formed posteriorly
by an osseous band, connecting the adjacent posterior edges of the peduncles of the two external trochlea,
a small oval foramen remaining above it for the transmission of vessels, and in front, as in all birds, by the
bridge connecting them anteriorly. The line of demarcation between the surfaces for the M. M. extensor
pollicis and adductor indicis, is imperceptible, and the inner limit of that for the M. extensor medii meets
thè outer at the lower extremity of the median concavity.
The inner border instead of being thin and ridge-like in its upper third, as in the Dodo, is replaced by
a flat plane with a sharp posterior edge, the anterior is rounded off in young individuals; this plane
slopes very slightly outwards, and terminates below at a rough projection situated at the junction of the
upper and middle thirds of the bone, corresponding to a minute one in the Dodo; beneath which, the
tumid inner part of the anterior aspect is broadly rounded off towards the internal surface. These surfaces
are separated by a prolongation of the posterior margin of the replacing plane, which descends to the metatarsal
facet, describing a curve convex posteriorly; the anterior margin of the plane is prolonged down on
the convexity of the anterior surface, at first parallel to, and afterwards converging to the posterior, meeting
it a little above the articular facet. The inner edge beneath this facet is less concave, being thinner than
in the Dodo, and more extended inwards. The medullary foramina have the same relative position with
regard to the shaft as in the Dodo. The fossa for the M.flexor brevis pollicis is narrower, from the replacement
of the inner edge, and more elongated, extending to within an inch of the metatarsal articular facet. The
edge of the calcaneal buttress is probably more convex. From the flattening of the outer border and the
less projection of the calcaneal buttress, the surface for the Abductor indicis is much narrower than in «the
Dodo, and passes more directly into the groove lodging the tendon, which, however, is deeper and more
distinctly defined than in the Dodo. The faint ridge which bounds this impression internally, and gives
attachment to the inter-muscular ligament from which the Flexor brevis pollicis arises, subsides before
reaching the articular facet.
The greater elongation and antero-posterior expansion of the central part of the shaft, and the breadth
of the outer and inner borders are the most characteristic and essential differences between the metatarsus of
the Solitaire and that of the Dodo. In all other respects they agree very closely ; the dimensions, even, of
the trochleas and of the upper extremity, and the absolute height of the posterior metatarsal articular facet are
alike in both. Those points in which the metatarsus of the Solitaire differs from that in the Dodo, are, in
some measure, repeated in Flaps, which has nearly the same relation to Geophaps that the Solitaire has to the
Dodo, in the proportionate lengths of the metatarsi. The metatarsus examined exhibits marks of disease
similar to those found in the bones of birds dying in menageries, the compact osseous tissue is opened out
along the lower moiety of the outer border, and in a circular space one inch beneath the proximal extremity,
so that the bone is more acted on by atmospheric agencies at these places; and a small .piece of the
lower node is removed (Plate XV. Fig. 2 a, 2 j |. The orifices of the minute periosteal Haversian canals
are more distinct than usual, and give the surface a granular aspect. The parietes of the shaft in the
immature specimen {ib. Fig. 4) are nearly one line thick; the medullary canal is divided, as usual, into-
three compartments by two thin partitions, which diverge as they pass from the anterior to the posterior
wall; the cancellated tissue extends ¡farther towards the middle of the shaft, in the narrow lateral,
than in the wide central division.
We have now ascertained that the cranium of the Solitaire resembles that of the Dodo in numerous
important points, differing in such respects only, as would justify us in regarding these birds as specifically
distinct. The metatarsus, also, is principally distinguished from that in the Dodo, by such variations in size
and proportion as might occur in species of the same genus. But in a small family, the members of which are
confined to distinct localities, we are warranted from analogy, in regarding each as forming the type of a
genus. The marked dissimilarity in external form between the Dodo and Solitaire, and the position of the
caruncular ridge in the latter, together with the shorter beak, fully justify the establishment of another genus
(Pezophaps) in thelidinee, to include this lost form. That the Dodo and Solitaire belong to the same extinct
sub-family of the CoUimbidce,. characterized chiefly by the peculiar structure of the cranium and rudimentary
condition of the wings, no one will, we trust, doubt, who has carefully and impartially examined the evidence
; the discovery of the osseous remains of the other extinct birds, supposed to belong to this group,
will enable us more strictly to define its boundaries, and its alliances with the other sub-families of the Order
Colmribce. We regard the Dodo, and its affine the Solitaire, as terrestrial’ flightless modifications of the Trero-
nine sub-type, but having no immediate affinity with the other ground Pigeons, as Goura, Calcenas, &c.,
which are more directly allied to the ordinary Columbines.
For the reception of that modification of the Treronine sub-type, represented by the Bidunculus, we
propose to establish the sub-family Gnathodontina, in the hope that other members of the group remain to be
discovered in the Polynesian Islands. The Gnathodontma are connected to the Treronma by the sub-genus
Toria, which differs from the typical Tr&ron in- the abbreviation of the mandibles, and in the pseudo-
raptorial form of the upper gnathotheca. The Bidunculus is essentially a perching bird, but terrestrial
affines probably exist, or have become extinct like the Dodo and Solitaire. The Pigeons form a perfectly isolated
group of birds, having no direct affinity either with the Insessores or Gallina. The rasorial genus Pterocles
approaches the Pigeons in the structure of the cranium, and in the form of the metatarsus; but it is destitute
of the peculiar columbine cere, and the hind toe, when present, is rudimentary and elevated. The
Gallina, then, approach the Pigeons through Pterocles, but no fusion of these groups is thus effected.
From other considerations, the Prince of Canino and Col. Sykes had, also, previously recognised the approximation
of Pterocles to the Columbida. The peculiar cere, and the great development of the nasal scales,
are characteristic of the Cohmbida, and probably have some relation to the mode in which the nestlings
are nourished. A milky fluid, analogous to the lacteal secretion in Mammalia, is elaborated by the thicks
ened mucous membrane of the crop of the parents, and poured into its cavity, where it mixes with the macerating
ingesta, and the young of certain species thrust their beaks into the throat of the parent, to obtain
the food thus provided.