of Humming-bird has the same character, whence the name "Androdon." The Dodlet (Didunculus), of the Samoan Isles, has been called the "tooth-billed Pigeon," because of the notches leaving three pointed horny processes in the sheath of the lower bill, beneath and just behind the hook-like production at the end of the upper one. The alveolar borders of the bill in Anatidæ and Phœnicopteridæ are notched by transversely set laminæ: and these are produced and pointed in their fish-catching allies, the Goosanders and Mergansers.
But in all these cases the "teeth" of the ornithologist or "tooth-like processes" are horny, are confined to the sheath of the bill, and there are no corresponding productions of the supporting bone, the alveolar borders of which are even, or but minutely indicative of the horny teeth. It is true, as Geoffroy St.-Hilaire first pointed out, the beginnings of the horny sheath are due in some birds (Parrots, e.g.) to detached papillæ occupying shallow cavities of the borders simulating sockets; but the primitive tubercles run into each other, and are ultimately confluent with the beak-sheath[1].
Perhaps a nearer approach to a dental structure is made where the hardening salts are in such excess as to give the sheath the character of ivory, welded to the bone, as in some Woodpeckers.
The production of the alveolar border into bony tooth-like processes is peculiar, according to my present observation of birds, to Odontopteryx. The closest repetition of this structure which I have yet seen is in the Australian Hooded Lizard (Chlamydosaurus); but the teeth are small, save the two at the fore part of each upper jaw and the single one at the same part of each mandibular ramus. The smaller teeth are so closely confluent with the alveolar border of both jaws as to seem to be processes: the larger anteriorly terminal teeth, though anchylosed to the bone, have their base defined by a ridge, suggesting the outlet of a socket, which is best marked in the lower jaw. All these teeth are tipped or capped with hard dentine; but such is not the case with the bony tooth-like processes in Odontopteryx. These seem, moreover, to have been sheathed with horn, or to have supported tooth-like processes of the horny beak; and their outer surface shows, though more feebly marked, the linear and punctate indentations relating to the vascular attachment of the horny to the bony beak. There is no trace of alveoli, although the cavity in the base of what seems to be a broken-off tooth at the fore part of the right upper jaw might be mistaken for one. I have not been able to detect, by application of lenses of any available power to the teeth in situ, any indication of a dentinal cap or apex.
After having myself outlined the drawings (which were finished as in figs. 1–6, Pl. XVI, with the care and accuracy characteristic of the accomplished artist, Mr. Griesbach) I had a mould and cast taken of the unique fossil to represent its original condition, and then selected the dental process which seemed best to promise evidence of tooth-structure.
Of this tooth (Pl. XVI. figs. 1 & 5, e) a longitudinal slice was taken (as in Pl. XVII. fig. 1) and laid, with some loss of the apex, upon a
- ↑ Anat. of Vertebrates, ii. p. 145.
