119. Hermasidas (Eermsea, Stiliger, Alderia).
20. Phy llobr anchidas (Phyllobranchus, Gyerce, Galiphylla).
21. Elysiidae (Flysia, Thu/ridilla, Placobranchus).
22. Limapontiidae (Limapontia, Genia, etc.).
A few cladoliepatic forms show an affinity to the Holohepatica. Such are Dirona and
Gharcotia in which the liver is trilobed but otherwise a solid mass. Their radula,
however, is of the specialized cladohepatic type.1 Doridoeides, on the other hand, -has the
appearance of a Dorid and triaulic genital ducts, but its alimentary system is distinctly
cladohepatic.
The Hedylidae are a family of very uncertain position. Nominally they consist of a
single genus Eedyle, of which Bergh has described one species from Flores, and
Kowalevsky3 four from the Black and iEgean Seas. But there are considerable
differences between the species, for whereas E. weberi Bergh has only two tentacles and
is definitely cladohepatic, Kowalevsky’s species have four tentacles (except E. milas-
chewitchii) and the liver is not ramified, though it also seems not to be a compact mass as
in the Holohepatica, but elongate and twisted. But in any case they differ from all
other known Nudibranchs. The visceral mass is distinctly marked off from the foot;
there are no branchial appendages; the skin is spiculous; there are no jaws and the
formula of the radula is 2 + 1 + 2. The genitalia are imperfectly known, but the verge
is armed with spines.
The remaining non-ascoglossan Cladohepatica may be divided into fifteen families,
viz. Pleurophyllidiidae, Dendronotidas, Scyllasidse, Bornellidse, Tethvmelibidas, Lomano-
tidaa, Phylliroidae, Janidas, Notaeolidiidse, ^Eolidiidse, Fionidas, Heroidae, Grlaucidae, Doto-
nidae, Myrrhinidae. I t would be satisfactory to reduce this list, for it contains two groups
of families corresponding roughly to Pelseneer’s Tritonioidea (Nos. 5—10 on the table)
and .ZEolidioidea (Nos. 13—18). But it seems natural and convenient to make the
iEolidiidas sensu stricto a family, and not to complicate the definition by including
forms which do not exactly correspond to it. The JEolidiidaa sensu stricto are probably
nearly as numerous as all the other Cladohepatica put together, and this being so the
type becomes important and even its minor peculiarities deserve attention. But if we
remove from the JEolids Doto, Fiona, Eero, etc., we cannot unite these forms into one
family for each has its own characteristics. Again, Families 3—10 on the above list
may very well be united into one group. But they also present marked differences,
and if we divide the asolidiform animals into six families it would be inconsistent to lump
these others in one. They must therefore, I think, be subdivided. But Families 5—10
and 13—18 may.be recognized as forming groups or sub-tribes, and if it is thought worth
while to give them names they may be known as Dendronotoidea and -dSolidioidea.
The Pleurophyllidiidse present such a singular combination of characters that they
must be regarded as a group by themselves. They are moderately large animals, tongue-
1 Dvrona 2.1.2 and Ghcurcotia 1.1.1. The affinities of Gharcotia are very uncertain.
2 “ Les Hedylidés, étude anatomique.” Mém. Acad. Pétersbourg (8) Phys. Math., xii, 1903. The
figure of Cylichna trwncatula given in Meyer and Möbius 1, p. 87, suggests that like Hedyle it has a
short foot and a visceral mass prolonged above and behind the foot, but as the animal has a shell it
is not possible to be sure of the size of the soft parts.
shaped and resembling the Holohepatica in their general appearance and bucdal parts.
They have jaws and a wide radula which is practically that of Tritonia, but the teeth are
denticulate. The rest of the digestive apparatus is as in the Solids, that is to say, it is
of the highly developed cladohepatic type. The liver is a layer of ramified tubes in the
body-wall which unite and enter the stomach by three ducts. Cnidosacs are present in the
mantle-margin and the branchiae are represented by lamellae under it. Pleuroplvyllidia
thus combines the buccal parts of Tritonia with the hepatic system of JEolidia and the
branchiae of Phyllidia. I t is a great difficulty to those who would make a phylogenetic
tree of the Nudibranchiata.
There are about fifty species divided between four or five genera, of which the most
important are Pleurophyllidia with branchiae and Pleu/roleura with none.
The six families grouped together as Dendronotoidea show traces of relationship to
Tritonia, but the back usually bears a few large papillae. The liver is usually contained
partly in the body-cavity and partly in these papillae, to which it sends branches, and there
is considerable variation in this respect even in the same species (especially in the genera
Dend/ronotus and Bornella). But even when there are no branches the liver is not as in
the Holohepatica, but is divided into three (more rarely two) parts and is also more
diffuse and flocculent. There is an oral veil, and the rhinophores are usually perfoliate
and retractile into tall sheaths. The radula is generally of moderate width.1 In Tethys
there are neither teeth nor jaws ; in Melibe no teeth but very feeble jaws. In this form
(and also in Scyllsea and Bornella) the stomach is strengthened by a hard armature and is
divided into two chambers. The arrangement of the hermaphrodite gland varies greatly,
depending on the.diffuseness of its own globules and that of the liver. In Scylleea and
Phylliroe it is divided into several (two to six) spherical packets. The male genitalia are
armed only in Bornella.
The external appearance of these animals is very varied. Dendronotus and Lomanotus
have somewhat the aspect of Tritonia, but the former has large arborescent cerata, and the
latter a row of simple cerata set on its undulating mantle-margin. Bornella is an exceedingly
elegant and active animal, somewhat resembling Eero in shape. I t has branched
appendages on the head and a few pairs of branched cerata which bear small gill-tufts.
Similar gill-tufts occur in Tethys and Scyllsea. Tethys and Melibe look like gigantic
.ZEolids with a great cowl or funnel round the mouth, which acts as a net to catch small
crustaceans, etc. Scyllsea is semi-pelagic, living on floating seaweed. I t has two pairs of
very large cerata. Phylliroe is truly pelagic and modified in accordance with its life. It
has no branchiae or appendages of any kind except rhinophores, and the liver is reduced
to four caaca which enter the stomach..
None of the six families are rich in genera or species, and when- several genera are
recognized in one family they are usually very similar. The families on the other hand
are very distinct, especially in external appearance, and may be tabulated as follows :
a. Pelagic: No cerata.
1. Phylliroidaa.
b. Not pelagic : cerata present.
1 In Scyllsea as much as 54.1.54, in Lomanotus 40.1.40, Bornella 19.1.19, Dendronotus 21.1.21
(but generally less). A few forms, such as Hancockia, have 1.1.1.