A revision of the Australian fossil species of Zoila (Gastropoda: Cypraeidae) 
3 
species of the two genera. This statement is incorrect, as shown 
by the descriptions and illustrations in Dolin and by my own 
observations of two of the species, Cypraeorbis arlettae and C. 
wilcoxi (see figs 3H-J, N, K, O). Revision of American taxa is 
beyond the scope of this paper, but I consider that Floradusta and 
Loxacypraea are synonyms of Cypraeaorbis, as the type species 
of each have a fossula similar to C .sphaeroides and C. 
ventripotens. Some of the other species included in Floradusta 
by Petuch (2004) are unlike species of Cypraeorbis in that the 
fossula is crossed by extensions of the columellar teeth and 
therefore cannot belong in the Cypraeorbinae. These species are 
best left in the genera to which they were assigned by Dolin 
(1991). 
Bernaya, type species Cypraea media Deshayes, 1835, was 
based on a species from Valmondois in the Paris Basin, Sables 
Moyens, Bartonian (late Eocene). The differences between 
Cypraeorbis and Bernaya are very slight. The fossula of 
specimens of Bernaya media (Deshayes) (figs 1A, 4B, F) from 
Bois du Roi, Auvers (MNHN), is large, subrectangular and 
concave. On the interior side, it is bounded by a prominent lateral 
ridge, in which there is a prominent anterior notch. The terminal 
ridge of the anterior canal extends into the interior of the aperture 
and merges with the lateral ridge of the fossula. A well-developed 
groove adjacent to the terminal ridge runs down into the fossula. 
This groove is partly obstructed by a columellar tooth, which 
merges with the terminal ridge. Specimens of Cypraea 
bartonensis Edwards, 1854 from Barton, Hampshire, England 
(NMV P310165-6) have a similarly structured fossula, but there 
are no columellar teeth obstructing the groove. In this respect, it 
closely resembles Cypraeorbis ventripotens. A specimen of 
Cypraea splendens Grateloup, 1827, from the Rupelian of Gaas, 
Landes, France (NMV P310168) also has a fossula identical to C. 
ventripotens. 
MacNeil and Dockery (1984) described the protoconch of C. 
sphaeroides as having five whorls, some of which are sculptured 
with a fenestrate pattern. Dolin and Dolin (1983) have described 
and figured (figs 3 and 4) the protoconch of Bernaya media as 
having four whorls, which have a fenestrate sculpture on the last 
two whorls. This form of protoconch suggests a planktotrophic 
larval stage. As there are no significant distinguishing features 
between the two genera, it seems appropriate to synonymise 
Bernaya with Cypraeorbis. 
The type species of Zoila is Cypraea friendii Gray, 1831, 
from the waters off Fremantle, Western Australia. Specimens of 
this species have a large spoon-shaped, deeply concave fossula 
(Wilson and Clarkson, 2004, pi. 47). The terminal ridge is not 
prominent, if at all present (on some specimens it is merely 
represented by a blunt thickening), and there is no prominent 
ridge bounding the edge of the fossula. There is also no groove 
running from the ventral surface down into the fossula, though 
there may be a gap between the labial teeth representing such a 
groove. There is no notch on the internal edge of the fossula. In 
these respects, it is unlike the American and European species; 
however, other species of Australian Zoila do have these features, 
but the edge of the fossula is not markedly notched, rather, 
merely indented (figs 1E-G). 
The fossula in Zoila chathamensis (figs 1H, 5I-J) is weakly 
developed, smooth, broad, very slightly concave, slightly 
projecting, and bounded by a very weak terminal ridge. It 
resembles the fossula in Z. didymorhyncha sp. nov. Specimens of 
Zoila viathomsoni sp. nov. have a well-developed fossula (figs 
ID, 5L) bounded by a prominent ridge and with a very weak 
indentation towards the anterior end. There is a very shallow 
sulcus running into the fossula. Specimens of Zoila platypyga 
(McCoy) have a fossula somewhat similar to Z. friendii but the 
edge of the fossula is more ridge-like. There is no sulcus 
extending into the fossula parallel to the terminal ridge. 
Specimens of Z. didymorhyncha sp. nov. (figs IE, 6C) have a 
weak terminal ridge merging with the edge of the fossula and a 
shallow sulcus running parallel to it into the fossula. None of the 
Australian fossils have the well-defined fossular margin or the 
well-developed notch behind it that is present in the European 
and American species of Cypraeorbis. 
Unlike the protoconchs of Cypraeorbis species, the 
protoconchs of species of Zoila, where known, are paucispiral, 
slightly tilted and smooth (Ranson, 1967, pi. 6, figs. 1-3; Wilson, 
1998). The protoconch of Z. platypyga consists of one whorl 
slightly tilted from the axis of the shell. The protoconch of Z. 
gigas (McCoy) consists of two whorls, the first of which is 
irregular and granulated (figs 13E-F). In cross-section (figs 
2A-C), the protoconchs of Z. viathomsoni, Z. platypyga and Z. 
gigas resemble cross-sections of protoconchs of Umbilia species 
(Darragh, 2002, fig. 1). This morphology suggests a protoconch 
formed from a partly chitinous whorl. Wilson (1985) has shown 
that living species of Zoila have direct development. The 
protoconchs of the fossils suggest that the fossils also had direct 
development and that the change from planktotrophic 
development to direct development had taken place by the late 
Eocene. Unfortunately, the protoconch of the earliest known 
species, Z. chathamensis, is unknown. Zoila species have direct 
larval development like the other Australian cowries referred to 
the genera Austrocypraea Cossmann, 1903, Umbilia Jousseaume, 
1884 and Notocypraea Schilder, 1927. The last three genera are 
endemic to Australia and do not have a distribution in shallow 
tropical seas, whereas species of Zoila are known from tropical 
as well as temperate waters and also have been found in the lower 
Tertiary of New Zealand, and the upper Tertiary of Indonesia and 
India. Austrocypraea, Umbilia and Notocypraea seem to be 
representatives of an ancient southern group, whereas the 
distribution of Zoila suggests that its origin may have been 
Tethyan. Most tropical cowries are known to have planktotrophic 
larvae, which accounts for their widespread distribution in the 
Indo-Pacific realm. Zoila seems to have evolved from the 
widespread genus Cypraeorbis, whose species almost certainly 
had planktotrophic larvae, although the occurrence of a species 
of Zoila in the Chatham Islands as early as the Paleocene to early 
Eocene may suggest a southern origin. 
Though the differences between the genera are slight, Zoila 
seems sufficiently different from Cypraeorbis to be maintained 
for the present. Wilson and Clarkson (2005) suggested that Zoila 
evolved in Asian waters from a directly developing Tethyan 
ancestor of the subfamily Bernayinae and migrated to Australia 
in Miocene time. However, the occurrence of species in Australia 
in the late Eocene and in the New Zealand region in the 
Paleocene/early Eocene suggests that Zoila evolved much 
earlier, probably from Cypraeorbis or some other closely related 
