_Corpus oblongo-ovale, compressum. Caput obtusum. Os parvum, haud
protractile. Dentes maxillares omnes incisores, parvi, æquales,
contigui, uniseriati; palatini velutini, minuti. Ossa suborbitalia
denticulata. Preoperculum margine adscendenti levissimè denticulato.
Operculum inerme. Membrana branchialis quatuorradiata. Pinnæ
verticales squamis confertis ferè omnino obtectæ: dorsalis unica,
subæqualis, membranâ ad apices spinarum parum laciniatâ: ventrales
radio primo molli elongato. Linea lateralis sub terminationem
dorsalis interrupta. Squamæ rostri et verticis parvæ; operculi et
corporis magnæ, obliquè dispositæ; omnes levissimè ciliatæ._
This apparently new form will enter into none of the genera established
by Cuvier and Valenciennes. The palatine teeth serve to detach it from
the _Sciænidæ_, while this character, taken in connexion with the
compressed body, and the extreme scaliness of the vertical fins, require
that it should be arranged with the _Chætodontidæ_, or at least have a
place in that large group to which Cuvier has given the name of
_Squammipennes_. It belongs to the second tribe in that family
characterized by cutting teeth; and it would seem most nearly allied to
_Pimelepterus_, but it does not approach that genus very closely, and
may at once be distinguished from it, by the teeth being without spurs
behind, and the dorsal and anal fins being more scaly. From
_Dipterodon_, the only other genus in that tribe, it may be known by its
undivided dorsal, independently of other marked differences.
But though this genus requires to be arranged with the _Chætodontidæ_ on
the grounds above mentioned, in all its other characters it comes much
nearer that portion of the _Sciænidæ_ which have the lateral line
terminating beneath the end of the dorsal fin; especially _Pomacentrus_,
which it resembles in the general form of the head and body,
denticulated suborbital and preopercle, unarmed opercle, four-rayed
branchiostegous membrane, and in the size and mode of arrangement of the
scales on the body. I am not aware that any species of _Pomacentrus_
have the dorsal and anal fins so completely covered with scales: but,
according to Cuvier and Valenciennes, there is a species of
_Glyphisodon_,[25] to which genus _Pomacentrus_ is closely allied, which
has these fins almost as entirely scaled, as in the true
_Squamipinnati_; and if so, there is nothing but the palatine teeth
which of necessity demands the separation of this new genus from the
_Sciænidæ_. These teeth can be distinctly felt upon the vomer, but I am
not quite sure from the small size of the fish, and its mouth also being
small, whether they exist on the palatines as well. It may be added that
this genus shews further itself an affinity to _Glyphisodon_, in the
filamentous prolongation of the first soft ray in the ventrals. This
character is not, I believe, found in _Pomacentrus_.
In which ever family it is placed, it forms a beautiful connecting link
between the two. It is from the Cape Verde Islands.