and hermaphrodite gland are diffused in My whine, more so it would seem than in the
Hermseidse. Although no iEolid presents seven principal ganglia in the nervous system
like those of the Ascoglossa, yet Rizzolia modesta is stated by Bergh to have a ganglion
on the visceral commissure, and this commissure bears ganglionic cells in other forms.
As explained in the preceding chapter, the statement commonly made in text-books that
all .ZEolids are diaulic is not correct. I t is probable that further research will show that
various forms of complete or incomplete triauly occur in both the Solids and Ascoglossa.
Therefore, I can see no reason in the present state of our knowledge for detaching the
Ascoglossa from the Cladohepatica.1
We can now examine a question already alluded to, namely, whether the Nudi-
branchs with ascoglossan radulas should be united with the two genera of Tebtibranchs
which possess the same peculiarity. These genera, Lophocercus and Lobiger (composing the
family Lophocercidse), have thin, globular shells partly covered by parapodia, a branchia of
somewhat abnormal form,1 2 a kidney composed of numerous chambers placed in the mantle
above the branchia, a liver contained wholly in the visceral mass and not ramified, a
nervous system consisting of six principal ganglia without gastro-cesophageal ganglia, and
a genital system which, according to the accounts of both Bergh and Pelseneer, does not
diverge materially from that of the Bullacea. There are two spermatothecas (as in many
Bullids), and the vas deferens is an internal tube (as in Aphis trum), not an external groove.
The buccal parts and oesophagus are similar to those of the nudibranchiate Ascoglossa,
and the similarity is particularly striking between Lobiger and some Phyllobranchidse,
where the conformation not only of the radula but of the ingluvies buccalis and the gland
on the oesophagus is almost identical.
The affinities of Lobiger and Lophocercus to the Bullacea are not, I think, denied.
Such forms as Newnesia8 and Diaphana are interesting connecting links* leading up to the
ascoglossan arrangement. Diaphana differs from other Bullids in having no jaws or
stomach plates and a triseriate radula. In Newnesia the structure of the alimentary canal
is similar, but there is a diverticulum on the oesophagus and the radula is reduced to a
single series of teeth. The teeth fit into one another but there is no ascus or ingluvies
buccalis. Were these two organs added, the alimentary canal would be almost exactly as
in Lobiger.
It might be difficult to discuss the resemblances of the Lophocercidas to the nudibranchiate
Ascoglossa if they depended on the details of the ramified organs in the
alimentarv and reproductive systems, for, as already hinted, the best authorities are not
unanimous in their elucidations of these complicated arrangements. But this tendency to
ramification and multiplication of glands is peculiar to the nudibranchiate section of the
Ascoglossa. I t has some analogies in other cladohepatic Nudibranchs, but none at all in
the Lophocercidae, and in fact constitutes one of the chief reasons for doubting their near
1 Bergh, although he detached the Ascoglossa from the Nudibranchiata for taxinomic purposes,
appears to have admitted that the Cladohepatica could be derived from the former. “ Zwischen den
Tectibranchien und den kladohepatischen Nudibranchien schieben sich als Bindeglied die Ascoglossen
ein.” System,-pp. 995—996.
2 I t is a strap bearing lamellae.
8 See Eliot in Journ. of Conchol., 1905, pp. 312—5.
relationship to the Nudibranchiata. The important points of resemblance seem to be (1)
the position of the genital orifices, (2) the nervous system, (3) the buccal apparatus, and
in some cases the structure of the oesophagus.
In all the nudibranchiate Ascoglossa as well as in the Lophocercidae, the male and
female orifices are separated by an interval and are not contiguous as in most Nudibranchs.
In the Lophocercidae this certainly points to a previous condition in which the two
openings were connected by a seminal groove (as in Newnesia), and the same explanation
may hold good in the other families. But in them not only are the male and female
orifices separated, but there are two female orifices which are also separated, sometimes
by a considerable distance. This arrangement cannot be regarded as anything but
extreme specialization, and taking the distribution of the three orifices as a whole it cannot
be said to be primitive or reminiscent of the Lophocercidae. On the other hand, in Fiona
(which is not ascoglossan) the male and female orifices are separate.
The nervous system of the Lophocercidae consists of six principal ganglia, and in the
nudibranchiate Ascoglossa of six or seven. The arrangement is usually not quite
symmetrical and appears to represent (1) a pair of cerebro-pleural ganglia intimately
fused together; (2) a pair of pedal ganglia; (3) three visceral ganglia of which one (e.g. in
Limapontia) may be suppressed. It is undoubtedly less specialized than the arrangement
found in other Nudibranchs, and it approximates the ascoglossan families to the Tecti-
branchs. But at the same time it may be noted that the gastro-oesophageal ganglia
characteristic'of Nudibranchs are absent in the Lophocercidae, but have been found by
Pelseneer in Gyerce, Hermaea, Elysia; also that the presence of a rudimentary visceral
ganglion in Rizzolia and in Doris approximates their nervous system to that of the
Ascoglossa.
In its complete form the ascoglossan radula is found only in the families we are
considering, but other molluscs have a somewhat similar arrangement of the buccal parts.
Besides Myrrhine and Newnesia already mentioned, the Prosobranch Homalogyra polyzona1
has a radula of uniseriate teeth fitting into one another, no jaws, and a buccal mass like
that of the Ascoglossa, although the worn-out teeth are not preserved. In Amphorina
coerulea and other Solids where the radula is long and tapering the teeth are preserved,
although there is no ascus. The small teeth (that is to say the early ones) remain on the
under side of the tongue, which is literally tongue-shaped and not a mere cushion. Only
those on the front part of the upper surface are used in eating, but the whole series
persists, so that the radula exhibits unusual length and unusual disparity of breadth. If
the under side of the tongue were enclosed by a membrane so as to form a pocket, this
would be an ascoglossan radula. Further, it should be remembered that the buccal parts
and in particular the radula may be almost identical in molluscs whose structure is
otherwise very different; the Prosobranchs Ianthina and 8'calaria have radulae very
similar to those of Dorids. Aldisa has teeth like those of Berthella (a Pleurobranchid),
and Ealinga (a phanerobranchiate Dorid) like those of Aplustrum. No information is
forthcoming as to the feeding habits of the Lophocercidae, but some of the nudibranchiate
Ascoglossa take nourishment by scratching or puncturing seaweeds and sucking the juice
1 The structure of this animal has been disputed, but recently has been investigated by Vayssiere,
Ann. des Sci. Nat., Ser. Zool., xix, pp. 363—377.