well preserved; but few other characters either external or internal can be established,
which is hardly surprising, as the specimens must be at least fifty years old. There is
no reason, however, to doubt the accuracy of Alder and Hancock’s description. The
dorsal margin is large, and seems to have borne in the one specimen six, in the other
eight branchiae of various sizes. No trace of stomach plates was found.
The jaws are yellow, rather long and narrow, and bear near the edge about four rows
of small prominences resembling a mosaic. The radula is very transparent, and consists
of twenty-five rows, which appear to contain thirty-six teeth on each side of the rhachis
when complete. The rhachidian tooth is tricuspid, and hollowed out below. The first
lateral is of the usual clumsy shape, but is rather variable in outline. The second lateral
is smooth, moderately stout, and simply hamate. The succeeding teeth become longer
and slenderer towards the outside. The third lateral bears a prominence or rudimentary
denticle, and the remaining laterals in the middle of the half row bear from
one to three long branch-like denticles, and sometimes one or two accessory shorter ones.
Towards the end of the row the denticles are found only at the tip of the teeth, and the
outermost are elongate and bifid. This peculiarity is not marked in Alder and Hancock’s
plate, which otherwise gives a very accurate representation of the radula.
DORIS L.
In compliance with the nomenclature introduced by Bergh, the common British Boris
is generally described as Archidoris tuberculata, and another form which is rare here
though frequent in French waters as Siaurodoris verrucosa. I submit however that this
nomenclature is not correct, and that the Linnaean name of Boris ought to be restored.
Bergh (Mai. Unt. in Semper’s Reisen, xiv, p. 616) decided to discontinue its use
(“ besser ware es, wie hier geschieht, den Namen Doris als generische Bezeichpung ganz
zu streichen ”), but there are at least two objections to this course. Firstly, if an old
genus is divided into several genera, one of these new genera should, according to the
rule generally recognized, bear the name of the old genus. Secondly, it would appear
that in Stmirodoris verrucosa Bergh, the use of the specific name really admits that the
animal is the Linnsean Boris. The type of Boris is Doris verrucosa of the tenth edition of
the ‘ Systema Naturae.’ It is true that the animal cannot be recognized from Linnaeus’s
description, but Cuvier identified it rightly or wrongly with a Mediterranean form, and
Cuvier’s animal has been renamed Stau/rodoris verrucosa by Bergh. But this form can
bear the specific name verrucosa only on the supposition that it is the Boris vermcosa of
Linnaeus. Therefore either it is Boris verrucosa, or else Staurodoris with a new specific
name ; but it cannot logically be Staurodoris vermcosa.
Further, it seems a pity to abolish a well-known name employed by so many eminent
naturalists, and in my opinion the use of Boris is not only correct but convenient. I
cannot help thinking that the distinctions between Bergh’s genera of the Archidorididae
are somewhat minute, and that a juster classification would be secured by the use of the
genus Doris (type Doris verrucosa), to include as sections or sub-genera at least
Stau/rodoris, Archidoris, Anisodoris, and possibly others.
Staurodoris Bergh cannot be satisfactorily separated from Archidoris Bergh, as the
two genera are connected by their less typical members. The typical Staurodoris has
simply pinnate branchiae and the back studded with clavate tubercles, which form valves
round the rhinophores and branchiae. But in the less typical form the branchiae become
bi- or tripinnate and the valvular tubercles less distinct.
The genus Doris as restored must consist of Doris verrucosa and such forms as can
be considered as belonging to the same genus. There may be considerable difference of
opinion as to how much these words include, but I would suggest for present use the
following definition and division.
Cryptobranchiate Dorids which are moderately soft and moderately flat. Dorsal
surface warty or tubercular. Tentacles often thick and grooved. Foot and mantle-
margin, as a rule, moderately broad. No armature on the labial cuticle or male genitalia.
Radula broad with no central tooth and numerous simply hamate laterals.
This definition is meant to exclude forms with a smooth or villous back, denticulate
teeth, and other peculiarities. A rudimentary armature of the labial cuticle seems to
occur occasionally, and it may be doubted if its presence in a fully developed condition
would justify us in excluding a species if otherwise possessing the generic characters.
The same may be said of the armature of the male genitalia.
Section 1. Staurodoris Bergh, which must also be considered as Doris sensu stricto.
Dorsal tubercles large and club-like, sometimes connected by ridges and specially
developed so as to protect the branchiae and rhinophores. Branchiae usually rather
scanty and often simply pinnate. Type: Doris verrucosa.1
Section 2. Archidoris Bergh. By reduction of the dorsal tubercles and amplification
of the brancliise Staurodoris passes imperceptibly into Archidoris, and forms on the
border line can be referred to either genus.2 The type of these forms with tripinnate
branchiae and moderately large tubercles which are not specially developed to protect the
branchise and rhinophores is Boris tuberculata Cuv.
Section 3. Gtenodoris Eliot. This exotic genus has most of the characters of
Archidoris but the branchial plumes are simply pinnate and arranged in a crescent. The
upper lip of the branchial pocket shuts down upon this crescent like a single valve. The
type is Doris fiabellifera Cheeseman (1880) from New Zealand.
Section 4. Anisodoris Bergh (synonym Montereina MacFarland) comprises exotic
forms which have the external characters of Archidoris and differ only in the presence
of a prostate on the male genitalia. Type Doris punctuolata.
Section 5. Homoiodoris Bergh has like Anisodoris the external characters of
Archidoris and a large prostate but also an armature of plates on the vagina. The type
Doris japonica Bergh is common in Japan.
1 A. and H. on p. xvii of the appendix to part vii suggest that D. verrucosa belongs to
Ehrenberg’s genus Glossodoris. But apart from the fact that this species is Linnaeus’s type for the
genus Doris, and therefoi*e not transferable from that genus, it is probable that Glossodoris is a
synonym for part of Ghromodoris.
8 E. g. Staurodoris pseudoverrucosa von Jhering has large bipinnate branchise and no tubercles
round the branchial pocket. Archidoris australis Bergh has “ nodules ” round the branchial and
rhinophorial pockets, and eleven slender branchise, simply pinnate below, bi- or tripinnate above.