Malacological Review, 1988, Suppl. 4: 205-240
PHYLOGENETIC RELATIONSHIPS IN THE GASTROPOD
FAMILY ARCHITECTONICIDAE, WITH NOTES ON THE
FAMILY MATHILDIDAE (ALLOGASTROPODA)
Rudiger Bieler
•.
ABSTRACT
A cladistic analysis for 12 genus-group taxa of the Architectonicidae (Architectonica, Philippia, Psilaxis, Basisulcata, Discotectonica, Granosolarium, Solatisonax, Pseudotorinia, Pseudomalaxis, Spirolaxis, Heliacus, and an unnamed new taxon), including
most of the Recent and fossil species, was conducted using the computer program
packages PAUP and PHYSYS. The family Architectonicidae is well-defined by a
suite of synapomorphies (e.g., a crescent-shaped osphradium, the lack of lateral
radular teeth, a projection on the inner side of the operculum). Possible outgroups in the
'Heterogastropoda' and Allogastropoda are discussed, and the data available for a
species currently grouped in the Mathildidae, Gegania valkyrie, were used for
outgroup comparison. Architectonicidae and Mathildidae, forming the superfamily
Architectonicoidea, share synapomorphies in anatomy (e.g., two juxtaposed ciliary
tracts on the left side of the mantle cavity; fused salivary glands; a longitudinal crest
at the ventral surface of the mantle cavity, built up by the anterior arterial vessel, a
so-called chordoid tissue and a posterior pedal gland). Radular and opercular characters thought to be synapomorphic for the Architectonicoidea by other authors are
discussed and rejected: the prominent peg on the inner side of the architectonicid
operculum is lacking in mathildids, and the "five-toothed-taenioglossate" radulae of
mathildids and most architectonicids probably evolved independently by loss of the
outer marginal or the lateral teeth, respectively. Twenty-six binary and three multistate taxonomic characters of anatomy, radula, operculum, protoconch and teleoconch
are selected and discussed. Special emphasis is placed upon sculptural features of the
teleoconch for which homology can be inferred from multiple lines of evidence. Additional autapomorphies are listed for architectonicid genera. The analysis resulted in a
resolved cladogram that showed three well-supported clades (Architectonica + new
taxon + Philippia + Psilaxis + Basisulcata; Discotectonica + Granosolarium + Solatisonax; Pseudotorinia + Pseudomalaxis + Spirolaxis). The position of Heliacus in the
cladogram is weakly supported; this genus is mainly defined by unique derived characters and retained symplesiomorphies. Three architectonicid genera known only from
fossils (Stellaxis, Nipteraxis, Climacopoma) can be added to the tree on the basis of
shell characters alone. The branching sequence of the cladogram supports the
hypothesis that in Architectonicidae smooth-shelled groups with oval-shaped
opercula and "ptenoglossate-like" radulae evolved later than sculptured forms with
circular opercula and "taenioglossate-like" radulae. Thus the "ptenoglossate-like"
radula of Architectonica must be interpreted as convergent when compared to mesogastropod families such as Epitoniidae and Janthinidae. Seven characters that display
homoplasy are discussed and hypotheses explaining parallelisms are offered. The
branching pattern of the cladogram is supported by fossil and ontogenetic evidence. The
use of the generic names Architectonica for the subgenera Architectonica s.s. + new
taxon, Philippia for Philippia s.s. + Psilaxis, and Pseudomalaxis for Pseudomalaxis s.s.
+ Spirolaxis is possible because each represent monophyletic groups. Earlier subdivisions into subfamilies are shown not to be based on monophyly and are rejected.
INTRODUCTION
For more than a century, the systematic position of the genera now comprising
the marine gastropod family Architectonicidae Gray, 1850 (= Solariidae Chenu,
1859) has been a matter of dispute. Features unusual for the mesogastropod prosobranchs, with which they were usually grouped, such as a hyperstrophic proto(205)
Rudiger Bieler
206
conch, an acrembolic proboscis, a "pigmented mantle organ" and atypical radulae,
led to various attempts to group the architectonicids with other proso- or opisthobranch families and to find a system to subdivide the family, which has a
wide spectrum of shell, radular and opercular characters. The proposed systems
differ greatly from each other, being introduced before sufficient taxonomic data ·
were available. A number of recent publications (Boss & Merrill, 1984; Haszprunar, 1985b,d,e; Melone & Taviani, 1985; Bieler, 1984a, 1985a,b, 1987) filled many
gaps in our knowledge of the group, but have not yet resulted in a satisfactory
classification above the generic level.
It is the scope of this paper:
- to define taxonomic characters based on homology in the family Architectonicidae;
- to discuss possible outgroups, and thus the relationships with other
families;
- to critically evaluate some of the presumed shared character states that
have been used to link the Architectonicidae with the family Mathildidae; and
- to present a phylogenetic analysis of the relationships within the Architectonicidae that allows the definition of monophyletic units and discrimination
between symplesiomorphies (shared ancestral character states), synapomorphies
(shared derived character states) and autapomorphies (derived character states
of terminal taxa not shared with other groups) in order to facilitate further discussion.
Phylogenetic analyses on a "lower" taxonomic level are still fairly scarce in
malacology. As a method that has been constantly changing and improved upon
over the past years, as philosophical discussions, mathematical algorithms and
computer technofogy are developing, a wide variety of different techniques and
approaches have been labelled "phylogenetic" or "cladistic" analyses. While
most authors nowadays present their data in a testable format, the actual step
from the data matrix to the cladogram presented is rarely documented and
therefore often not reproducible. (This is also the case when an author actually
cites the name of the computer "program" used. Most of the modem "phylogenetic
analysis programs" available are extensive packages comprising a variety of algorithms, sometimes ranging from phenetic to cladistic approaches.) In this
paper special emphasis is therefore placed on the methods employed.
MATERIALS AND ME1HODS
The taxonomic data for this study are based partly on previously unpublished
personal observations, partly on my recent revisions of the family Architectonicidae (Bieler, 1984a,b, 1985a,b, 1987), and partly on data from other authors, as acknowledged in the text and tables. Sources of material in addition to those
mentioned in the acknowledgements (and methods of measurement) are those
outlined in my publications cited. The terminology of architectonicid shell
sculpture is explained in Figure 1. While the names of architectonicid taxa are
those as recently revised, all names of Mathildidae mentioned in this publication are those used by the authors cited in the text. Generic placement and
specific identity need verification. [The family is currently under revision
(Bieler, in prep.). Following a conservative approach, Mathildidae s.l. is used,
....
Architectonicidae Phylogeny
207
including the genus Gegania Jeffreys, 1884.]
"Hennig's method" of phylogenetic systematics (Hennig, 1950, 1966), a
cladistic analysis that defines monophyletic groups based on internested sets of
synapomorphies (shared derived characters), was used in this study. [For an in-
umrセAZzNRコュᄋ@
SSR
..セᄋ@
.. セᄋオ@
I
...セ@ ..セᄋ@
penultimate whorl
UPR
suture
body whorl
UPR
LPR
peripheral keel
BF
II
base
UC
umbilicus
FIG. 1. Diagrammatic representation of the pattern of spiral ribs on the Architectonica
teleoconch, which can be homologized with spiral ribs in other genera [modified after Bayer
(1940: fig. 1) and Bieler (1984b: fig. 1)). Numbers refer to Bayer's (1940) system. Abbreviations in brackets are those used in earlier publications in the German language: SSR = subsutural rib [ssR]; UMR =upper mid-rib [oMFRJ; I.MR= lower mid-rib [uMFR]; MR= mid-ribs
(MFR]; UPR =upper peripheral rib [oRRJ; LPR =lower peripheral rib [uRR]; IPR = infraperipheral rib [ipRJ; PR = peripheral ribs [RR]; BF = basal field [BFI]; PUR = proxumbical rib
[puR]; UC= umbilical crenae [er].
troduction to the recent discussion of various methodologies see, for instance, von
Vaupel Klein (1984).]
The number of taxa and characters involved did not allow the construction of
cladograms "by hand" without the intuitive weighting of certain characters. Of
the several computer programs available for cladistic analysis, two packages,
PAUP (Swofford, 1985) and PHYSYS (Mickevich & Farris, 1984), were used.
The PAUP program and the WAGNER.S option of the PHYSYS package are
Rudiger Bieler
208
based on the "Wagner method" (Kluge & Farris, 1969; Farris, 1970), where the
preferred tree is the one of minimal length in a Manhattan metric (which is a
function of the range of character states of all taxa in the study), with no a priori
restrictions on the nature of permissible character changes. Of the phylogenetic
methods, that of Wagner shows greatest stability, and is less sensitive to the
amount of homoplasious (parallel/ convergent) evolution than phenetic methods
(Mickevich, 1978). Both programs have several major advantages over earlier
programs (e.g., WAGNER78, WAGPROC) that also employed "Wagner
methods": (1) they are able to handle missing (not applicable or unavailable)
data; (2) they can treat "unordered" multistate characters [those for which the
transformation series is not known prior to the analysis; PAUP uses an algorithm
based on Fitch's (1971) optimization; PHYSYS uses Mickevich's (1982)
"Transformation Series Analysis"]; and (3) they can detect the existence of
multiple, equally parsimonious character-state reconstructions (incorporating the
least number of character- state changes). PHYSYS can handle non-linear
(branching) multistate characters, while PAUP requires recoding into several
binary characters for this purpose.
The data set used in this paper was small enough to allow the use of an exact
algorithm (one that will find all equally parsimonious trees). The algorithm
used for this purpose, the ''branch-and-bound-method" [as suggested by Hendy &
Penny (1982), modified by Swofford (1985) as part of the PAUP package], represents the most efficient approach currently available (see Platnick, 1987).
Of the several available rooting algorithms, outgroup-rooting was used: the
taxon designated as the outgroup forms a sister group to the in-group taxa; the
program "decides" polarities based on the same parsimony criterion being used in
constructing the tree (in contrast to earlier methods, e.g., "hand cladistic" and
WAGNER78, where polarity decisions were made in a separate step prior to tree
construction).
The terms monophyly and paraphyly are employed in their usual meaning,
i.e., according to the concepts of Hennig (e.g., 1966: fig. 45) and as defined by
Farris (1975). Further details are outlined in the sections "Selection of Characters" and "Tree Construction" below.
The abbreviation "Fig." refers to figures in this work; "fig." refers to those in
other publications.
Cited repositories are:
AMS
LACM
NMNZ
SMF
USNM
Australian Museum, Sydney
Los Angeles County Museum of Natural History
National Museum of New Zealand, Wellington
Senckenberg-Museum, Frankfurt/Main
National Museum of Natural History, Smithsonian Institution, Washington.
ANALYSIS
Selection of in-group taxa
The architectonicid genus-group taxa are as revised by Bieler (1984b, 1985a,b,
1987). For the purpose of this analysis subgenera as well as genera have been used
as operational taxonomic units in the cases of Philippia/ Psilaxis and Pseudomalaxis / Spirolaxis, while Heliacus s.l. was used, comprising all its recognized
)
!.
Architectonicidae Phylogeny
209
subgenera (see below). Basisulcata, the type species of which was previously
included in Philippia (Bieler, 1985a: 336), is here recognized as a genus-group
taxon. Genera known only from fossils, e.g., Climacopoma Fischer, 1885 (=
Patulaxis Dall, 1892); Dinaxis Dall in Aldrich, 1895; Eosolarium Chavan, 1947;
Nipteraxis Cossmann, 1915, and Stellaxis Dall, 1892, were excluded from the
analysis for lack of comparative data, but were subsequently assigned positions in
the resulting cladogram. Twelve architectonicid genus-group taxa were used in
the study, comprising almost all of the approximately 130 Recent species
worldwide (Bieler, 1986):
1) Architectonica Roding, 1798 (= Solarium Lamarck, 1799; = Verticillus Jousseaume, 1888, non Moquin-Tandon, 1848).
2) new taxon (Bieler,1987) (see note at end of manuscript).
3) Philippia J. E. Gray, 1847.
4) Psilaxis Woodring, 1928.
5) Basisulcata Melone & Taviani, 1985.
6) Discotectonica Marwick, 1931 (= Acutitectonica Habe, 1961; = Russetia
Garrard, 1961 ).
7) Granosolarium Sacco, 1892 (= Solariaxis Dall, 1892; = Claraxis Iredale, 1936).
8) Solatisonax Iredale, 1931. [The type species of the recently described taxa
Heliacus (Granoheliacus) Melone & Taviani, 1985, and Heliacus (Redivivus)
Melone & Taviani, 1985, appear to belong to Solatisonax, but need further
study and are here excluded from the analysis].
9) Pseudotorinia Sacco, 1892; (= Awarua Mestayer, 1930; = Calodisculus Rehder,
1935; ?= Punjabia Eames, 1952).
10) Pseudomalaxis Fischer, 1885 (= Discosolis Dall, 1892; = Mangonuia Mestayer,
1930).
11) Spirolaxis Monterosato, 1913 (= Paurodiscus Rehder, 1935; = Aguayodiscus
Jaume & Borro, 1946).
12) Heliacus Orbigny in Sagra, 1842 (= Torinia J. E. Gray, 1842) [comprising the
following subgenera: Heliacus s.s.; Torinista Iredale, 1936 (= Astronacus
Woodring, 1959); Grandeliacus Iredale, 1957; Teretropoma Rochebrune, 1881;
Gyriscus Tiberi, 1867; new taxon (Bieler, 1987)].
Systematic Position Of The Architectonicidae:
"Heterogastropoda" Versus Allogastropoda
The family Architectonicidae has received considerable attention in the past,
focusing mainly on two problems of higher classification: (a) the position of this
group in the Gastropoda, and (b) the composition of the group of related families
around Architectonicidae. The family, classically placed in the mesogastropod
superfamily Cerithioidea, has been variously raised to superfamilial level,
either without (e.g., Korobkov, 1955: 136) or with (e.g., Taylor & Sohl, 1962: 14)
the inclusion of other families, usually on the basis of shared radular features or
overall shell similarity. Thiele (1925: 113) pointed out the similarities in heterostrophy, and in opercular and radular characters between Architectonicidae
and Mathildidae, and suggested a phylogenetic relationship. Taylor & Sohl
(1962: 14) added the Omalaxidae to this group, and Golikov & Starobogatov
(1975: 223) included the Cyclostremellidae. However, the Omalaxidae, a caenogastropod family of uncertain position known only from the Eocene, should be
210
Rudiger Bieler
excluded from this group (see Bieler, 1984b: 65), because it does not share any homologous characters that could indicate close relationship with the other
families in question; Cyclostremella, type genus of Cyclostremellidae Moore,
1966, has been shown to belong to the Pyramidellidae (Robertson, 1973).
The "special" position of the Architectonicidae in the Gastropoda was
already pointed out by MacDonald (1860: 76), in his excellent but poorly known
work on gastropod classification. He placed Architectonicidae (as Solariidae,
including "Scalariil" = Epitoniidae in the modem sense) and Janthinidae close to
the Pyramidellidae, and suggested that this group might "form a remarkable
and beautiful link between the Opisthobranchiata and the unisexual Gastropoda." MacDonald thus outlined an early precursor of the "Heterogastropoda," a
taxon formally introduced more than a century later by Habe & Kosuge (1966: 101,
as an order) and Kosuge (1966: 320, as a suborder), for the families Architectonicidae, Epitoniidae, Janthinidae, Mathildidae and Triphoridae. Habe & Kosuge's
taxon has been accepted by a number of authors, comprising some or all families
originally included, as a subdivision of either the Proso- or Opisthobranchia
(e.g., Shuto, 1969; Cemohorsky, 1972, 1978; Oimo, 1975; Powell, 1979; Boss, 1982,
Fretter & Graham, 1982; Kilburn & Rippey, 1982; partly summarized in Haszprunar 1985a: tab.5). [For a discussion of prosobranch and opisthobranch traits in Architectonicidae, see Merrill (1970: 271) and the detailed account by Robertson
(1974); for opisthobranch-like characters of shell-matrix and spermiogenesis, see
Ghiselin et al. (1967: 13) and Healy (1982: 197), respectively.]
Robertson (1985) compared some of the characters of Architectonicoidea, Epitonioidea and Pyramidellidae with those of other prosobranch and opisthobranch families. Of the four characters listed by him, that "support the ideas
that, within themselves, the characters are homologous and reflect phylogeny"
(1985: 1), only one, the "pigmented mantle organs" of the larvae and postlarvae,
is shared by architectonicids with Epitoniidae and Janthinidae. The majority of
the characters shared between Architectonicidae and Epitoniidae, as listed by
Robertson (1974: 216), can be better explained by convergence due to similar diets
[i.e., association with cnidarians: cuticularized esophagus, ptenoglossate-like
radula (see below)] or as symplesiomorphy (streptoneury). The same is true for
corresponding similarities between Architectonicidae and Janthinidae. The
other characters studied by Robertson (1985) (i.e., the longitudinal ciliated strips
dorsal and ventral in the mantle cavity of postlarvae, the chalazae connecting
egg cocoons, and the hyperstrophically coiled larval shells) are restricted to
three families, Architectonicidae, Mathildidae and Pyramidellidae, forming a
group within the "assuredly polyphyletic" (Boss, 1982: 996) assemblage that has
been called "Heterogastropoda." The "ciliated strips," listed as "probably
absent" for Mathildidae by Robertson (1985: fig.2), do occur in that family (Haszprunar, 1985e: 202; pers. observ.).
Studying the fine structure of osphradial epithelium, Haszprunar (1985a,b)
arrived at an almost identical grouping: Triphoridae and Epitoniidae (as well as
Eulimoidea and Cerithiopsoidea which had been included in the heterogastropods by some authors) were shown to have an osphradial fine-structure characteristic of higher Caenogastropoda, while Architectonicidae and Mathildidae
displayed a different pattern.
Because of the inherent polyphyly of the "Heterogastropoda," Haszprunar
セᄋ@
.
Architectonicidae Phylogeny
211
(1985c: 25) introduced the new superorder Allogastropoda, to include three Recent
families: Architectonicidae and Mathildidae in the Architectonicoidea, and Pyramidellidae in the Pyramidelloidea. This arrangement was based on anatomical characters thought to be synapomorphic, i.e., the existence of a secondary foliobranch gill to the right of the dorsal ciliary tract, the acrembolic proboscis,
and the occurrence of spermatophores. He also included five fossil families: a
family named "Streptaclidae" (Haszprunar, 1985c: 25) or "Streptaclididae"
(1985c: 32) [err. pro Streptacididae Knight, 1931], which he placed in the Pyramidelloidea (but which he later excluded; Haszprunar, 1985e: 211), and the extinct
Nerineoidea, comprising four families [Nerineidae, Nerinellidae, Ceritellidae
and "ltieridae" (err. pro ltieriidae Cossmann, 1896)]. Of the several names
available for higher taxa comprising some or all of these families (see Ponder &
Waren, 1988, this volume), the name Allogastropoda Haszprunar, 1985, is here
used to describe the clade that is formed by at least the three extant families
mentioned, namely Architectonicidae, Mathildidae and Pyramidellidae (see
also Bieler, 1985b). [Haszprunar (1988, this volume) changed his concept of Allogastropoda drastically; he now views this group as an "orthophyletic grade,"
comprising five Recent superfamilies.]
.•
Selection Of Outgroup
The selection of an outgroup for the cladistic analysis of architectonicid
genera is problematic, because the choice of closely related groups with comparable characters is severely limited. Because this analysis involves a considerable
number of non-shell characters, only a family with Recent representatives is appropriate, as opposed to extinct groups for which no direct anatomical information can be collected. [Haszprunar (1985c), however, inferred for the fossil Nerineoidea a special secondary gill (1985c: 15), an acrembolic proboscis (1985c: 30)
and atypical sperm (1985c: 24); the last probably an error for Neritoidea, judging
from the reference cited (Giusti & Selmi, 1982).]
Of the three Recent families in the Allogastropoda, the Pyramidellidae are
often smooth-shelled, highly specialized ectoparasites which lack a radula
(Fretter & Graham, 1949; Fretter et al., 1986), and thus offer few characters for
comparison with the shell and radular data available for Architectonicidae.
Similar problems arise for the Omalogyridae, a family of incertae sedis that has
been tentatively grouped close to the Architectonicoidea/Pyramidelloidea
complex (Haszprunar, 1985c: 25; Salvini-Plawen & Haszprunar, 1987: 760). The
minute omalogyrids lack many of the organs and structures that are developed in
architectonicids (Fretter, 1948).
Thus, the only closely related family that displays a sufficient number of
comparable characters and could qualify as an outgroup on the basis of close phylogenetic relationship appears to be the Mathildidae. This small, deep-water
family had already been grouped with Architectonicidae by Thiele (1921: 143,
1925: 113), based on similarities in radular features and the common occurrence of
heterostrophy. Additional anatomical evidence for a phylogenetic relationship
was published by Climo (1975) and Haszprunar (1985e), who described a number
of synapomorphies in anatomy (e.g., a longitudinal crest at the ventral surface of
the mantle cavity, built up by the anterior arterial vessel, a so-called chordoid
tissue and a posterior pedal gland; fused salivary glands; two juxtaposed ciliary
212
Rudiger Bieler
tracts on the left side of the mantle cavity).
However, two of the features that were used in the past to combine the two
families need re-evaluation: the presumed shared characters in opercula and
radulae.
Opercula: All architectonicid opercula are characterized by a peg-shaped
internal projection (Fig. 2), where the muscle is attached. A close similarity of a rchitectonicid and mathildid opercula has been stressed in the literature (e.g.,
Thiele, 1925: 113, 1928: 88; Merrill, 1970: 274; Hasz prunar, 1985d: 41), but the
existence of this structure has not been d emonstrated for any mathildid species [it
is, however, present in other families, e.g., Triphoridae (Marshall, 1983: 4, fig.
ld)] . The opercula of nine mathildid species were examined (species-level systematics under revision; Bieler, in prep.), and they all lack an internal projection
(for example, see Fig. 3). The common possession of a corneous multispiral
operculum hardly qualifies as a synapomorphy.
Radulae: Only a few data on mathildid radulae are available. Climo (1975:
fig.3) illustrated (see Fig. 4) the radula of Gegania valkyrie Powell, 1971. This
radula has a pair of reduced outer marginal teeth (Fig. 4d) which form "an anchor
to which the bases of inner marginal and lateral teeth abut" (Climo, 1975: 277).
FIGS. 2-3. Opercula of Heliacus and Mathilda. 2. Oblique side view of operculum of
Heliacus tlariegatus (Gmelin, 1791), showing prominent peg on body side (SMF 256388). 3.
Oblique side view (body side) of opcrculum of Mathilda cf. quadricarinata (13rocchi, 1814)
[USNM 131944).
The tooth arrangement and cusp configura tion (Fig. 4a-c) implies derivation from
a mesogastropod-like taenioglossate form.
The radulae of Mathilda quinquelirata Kuroda, 1958 (see Fig. 5), and Eucharilda sinensis (Fischer, 1867) (see Fig. 6) as illustrated by Habe (1958: 56, pis. 2-3)
likewise have a smaller latera l tooth and a single, larger, differently-cusped
marginal tooth per half row (Fig. 5b-c, 6b-c). Outer marginals h ave not been
found. Habe's drawing of the E. sinensis rachidian (Fig. 6a) contradicts his text
(1958: 56) which states tha t the rachidian "has nine cusps normally," falling
within the range of the o ther species studied to date. Thiele's drawing (1928:
fig.10; see Fig. 7a-c) of the radula of Mathilda elegantissima (0.G. Costa, 1861)
reveals the same five-teeth-per-row configuration. Climo (1975: 280-281) misinterpreted Thiele's figure as being typically seven-too thed taenioglossate,
because he understood the lateral view of the rachidian (Fig. 7a') as the lateral
tooth, thereby counting seven teeth per row. Thiele's text (1928: 88) ("die Mittelplatte ... und die beiden Seitenplat ten") a nd fi g ure caption (1928: 87)
("Mittelplatte von oben und von der Seite gesehen"), however, indicate clearly
that Thiele observed only five teeth. Merrill (1970: pl. 8 fig.Id) again found five
Architectonicidae Phylogeny
rachidian
4
セ。@
213
lateral
inner marginal
outer marginal
c
5
c
-
6
c
-
c
-
c
-
.•
8
FIGS. 4-8. Mathildid radulae. All figures redrawn and newly arranged (original scale
kept within each figure). The letter "a" indicates additional view of rachidian: 4. Gegania
valkyrie Powell, 1971 [after Climo, 1975: fig.3]; 5. Mathilda quinquelirata Kuroda, 1958 [after
Habe, 1958: pl. 3 fig. 14); 6. Eucharilda sinensis (Fischer, 1867) [after Habe, 1958: pl. 2 fig. 11);
7. Mathilda elegantissima (0.G. Costa, 1861) [after Thiele, 1928: fig. 10); 8. Mathilda quadricarinata (Brocchi, 1814) [after Merrill, 1970: pl. 8 fig. ld).
214
Rudiger Bieler
teeth per row for Mathilda quadricarinata (Brocchi, 1S14) (see Fig. S). In his
drawing he placed the longest tooth (Fig. Sc) closer to the rachidian than the
tooth here reproduced in Fig. Sb. A comparison with the laterals and marginals
of the other studied species suggest that this is an artifact.
Ponder (19S5: fig. 145H) illustrated the radula of Brookesena neozelanica
(Suter, 190S), a species provisionally included in the Mathildidae. The family
Mathildidae is rather homogeneous in shell characters: all members have a
smooth hyperstrophic protoconch, except for Brookesena Finlay, 1927 (based on
Mathilda neozelanica Suter, 190S), which has a sculptured homeostrophic protoconch (Ponder, 19S5: fig. 145F). A similar protoconch (and teleoconch) can be
found in members of Turritellopsis G.O. Sars, 1S7S. However, Turritellopsis
acicula (Stimpson, 1S51), type species of that genus, is characterized by a
different radular type, with only three teeth per row (Sars, 1S7S: pl. 7 fig. 2),
and should not be included in the Mathildidae. Ponder's (19S5: fig. 145H) SEM
photomicrograph of Brookesena neozelanica shows a five-toothed radula similar
to those in Figures 5-S, the strong lateral having five finger-shaped cusps and a
wide, asymmetrical base forming an outer tongue.
An SEM photomicrograph of the radula of Mathilda cf. quinquelirata Kuroda,
1958 (see Fig. 9) shows a gap between the rachidian and laterals and a very
strong central cusp on the rachidian. The radula of a specimen identified as
Mathilda decorata Hedley, 1903 (by comparison with the holotype, AMS
C.16299; type species of Opimilda Iredale, 1929), has a rachidian with numerous
(w24) long, filiform cusps of subequal strength, the central ones being longer (Figs.
10-11). As in the other species studied, the inner marginals are longer and bear
more denticles than the laterals (Fig. 11). No outer marginals could be traced.
Rachidian and laterals are separated by a wide space as in M. cf. quinquelirata.
A character not seen in other mathildid species studied is a feather-like
extension on the marginals ("F' in Fig. 11), projecting at an oblique angle from
shortly below the tip of the "normal" tooth. This structure is highly flexible, but
breaks off easily during preparation and is not readily visible in light microscopy
under lower magnification.
The eight mathildid species (spanning several nominal genera) for which the
radulae are known, display much less variation in number and relative size of
teeth than previously assumed. The mathildid radula is modified taenioglossate, with the rachidian separated by a gap from a pair of laterals, which are
longer than the rachidian and bear a moderate number of cusps, and a pair of
multi-cusped or serrated inner marginals, which are longer than the laterals; the
outer pair of marginals is either greatly reduced or lost entirely.
This "five-toothed" radula of mathildids has been homologized with the
five-toothed radula in a number of architectonicid genera such as Heliacus, Pseudotorinia and Philippia. However, the taenioglossate-like radula of Architectonicidae is of a different structure and, probably, origin. The two outer pairs of
teeth are of equal length and structure (Fig. 12). No evidence of an additional
outer pair of teeth (as in Gegania) was found. The two outer pairs present closely
resemble, and are here homologized with, the inner and outer marginals in a
"typical" seven-toothed taenioglossate radula [see, for instance, the radula of
Cerithidea costata (da Costa, 1778) in Bandel, 1984: 37, pl. 3 fig. 6)). If this fivetoothed radula was derived as in the Mathildidae, by loss of the outer
セN@
Architectonicidae Phylogeny
215
FIGS. 9-1 4. 9. Radula of Math ilda cf. quinquelirata Kuroda, 1958 [AMS C.146527] (SEM
courtesy of Dr. Anders Wa ren). 10. Rachidian and lateral teeth of the radula of Mathilda
decorata Hedley, 1903 [AMS C.14868) (SEM). 11. Lateral teeth and inner marginals of the
radula of Mathilda decorata Hedley, 1903 [AMS C.14868 ). "F" indica tes feather-like extensions on inner marginal teeth (SEM). 12. Radula of T-leliacus variegatus (Gmelin, 1791) [SMF
256388) (SEM). 13. Radula of Arcliitectonica picta (Philippi, 1849) [LACM 78-65) (SEM). 14.
Protoconch of Gegania valkyrie Powell, 1971 [NMNZ M.36712] (SEM).
marginals, two evolutionary s teps would be n ecessary: (a ) the reduction of the
outermost pair of teeth, and (b) the transformation of the lateral teeth into
virtual duplica tes of the inner marginals. It is therefore proposed that the fivetoothed taenioglossa te-like rad ulae of Mathildidae and Architectonicidae
evolved independently, probably in parallel from a seven-toothed taenioglos-
Rudiger Bieler
216
sate ancestor. While in the Mathildidae the outer marginals were reduced, the
Architectonicidae lost the lateral teeth (Fig. 15). "Five-teeth-per-row" is not a
synapomorphy of the Architectonicoidea, but rather an autapomorphy of each of
the two families.
Data other than shell characters are available for only a few mathildid
species. The anatomy of only two species, Opimilda maoria Powell, 1940, and
Gegania valkyrie Powell, 1971, have been studied in detail (Climo, 1975; Haszprunar, 1985e; pers. observ.). Within the Mathildidae, G. valkyrie displays a
-
Mathilda
rachldian
Heliacus
lateral
inner
marginal
outer
marginal
-
FIG. 15. Diagrammatic figure showing hypothetical independent origin of "5-toothed
taenioglossate-like" radulae in Mathildidae and Architectonicidae.
number of ancestral character states in shell, radular and anatomical (especially
buccal system) characters and is therefore here used as the outgroup, rather than
0. maoria, which displays many derived character states within mathildids
(Bieler, in prep).
Selection of characters
Characters useful for phylogenetic analyses have to meet certain criteria:
they must be based on features thought to be homologous, they must display at
least two different character states, and the character states must be known for at
Architectonicidae Phylogeny
217
least three of the taxa involved. The character states used for terminal taxa in
an analysis on the generic level could be (a) that of the type species, (b) true for
some, but not necessarily all, species, or (c) true for all species in the genus. In this
study, the last approach was used, with a few exceptions explained in the text.
The inclusion of autapomorphies in cladistic analyses has meaning only as
long as they provide additional information that can be used to infer branching
sequence. This is the case when the autapomorphy is part of a multistate
character. However, character states of binary characters that are unique to one
of the taxa involved and have the opposite, plesiomorphic state in all others do
not supply this information, should not be used, and were therefore eliminated
from this analysis. However, these autapomorphies do serve to define terminal
taxa and might become useful for a phylogenetic study when additional data
become available, and are listed here:
Architectonica: presence of a distinct color pattern on head; rachidian absent
(some species); very large teleoconch size; color pattern of spiral bands on
shell.
Philippia: flanking cusps on rachidian as strong as central cusp; on protoconch,
callus overlays anal keel and false umbilicus.
Psilaxis: opercular peg mushroom-shaped.
Discotectonica: umbilical crenae flattened.
Granosolarium: umbilical crenae sunken into umbilicus; nodules on lower peripheral rib developed into spines (some species).
Solatisonax: upper peripheral rib on early whorls prominent and undulated.
Pseudotorinia: protoconch paucispiral (some species).
Spirolaxis: shell with open coiling; strap-like jaws.
Heliacus: diffuse black body pigment (some species).
In the following, the 26 binary and three multistate characters and their
states are listed. The code used in the analysis is added in parentheses [Characters 2 and 5 were inferred from Haszprunar (1985d,e), who did not explicitly
define size classes]. It should be noted that at this point no statement is made
regarding the direction of change in a series of character states (with one
exception; see character 15, below); the polarity decision is made by outgroupcomparison during construction of the tree, i.e., during the program run (see
"Materials & Methods"). The resulting data matrix is shown in Table 2
(Appendix).
,.
Anatomical characters
Alimentary tract:
1) buccal mass: developed (0), absent (1)
2) buccal mass: large (0), medium (1), small (2)
3) esophagus: simple (0), with blind sac containing "rod-like structure" (1)
4) esophagus: straight (0), looped (1)
5) proboscis sheath: short (0), long (1)
Other anatomical characters:
6) osphradium organization: bilamellar (0), crescent shaped (1)
The majority of the information available on architectonicid anatomy (i.e.,
Bouvier, 1886a,b, 1887; Risbec, 1955; Merrill, 1970; Robertson, 1974; Climo, 1975,
218
Rudiger Bieler
Haszprunar, 1985b,d) is useful for comparison between families, but does not
enable a comparative study within the Architectonicidae. Detailed information
on the anatomy of the architectonicid reproductive system is restricted to a few
species of Heliacus and Architectonica (Haszprunar 1985d: 33; pers. observ.).
Variation within the taxa, at both the specific and generic levels, is not yet
known, and the present data on reproductive strategies show amazing heterogeneity within Architectonicoidea [protandric and simultaneous hermaphroditism
in species of Heliacus and Discotectonica (Haszprunar, 1985d: 33, 34, 38), simultaneous hermaphroditism in Gegania, gonochorism in species of Philippia, Granosolarium and Opimilda (Haszprunar, 1985e: 205, 211)]. This probably accounts for
conflicting statements in the literature which state that the Architectonicidae
are hermaphroditic (Robertson, 1974: 218; Gosliner, 1981: 207) or gonochoristic
(Gosliner, 1981: 209; Boss, 1982: 997). As pointed out by Robertson (1985: 16), simultaneous hermaphroditism as a taxonomic character should be used with care
in phylogenetic studies, especially because other gastropod families are known in
which hermaphroditism and dioecism co-occur in a single genus (e.g., Patella;
Webber, 1977: 11). Reproductive data are considered too fragmentary to be
included here.
Radular characters
7) radula: present (0), lacking (1)
8) no. of flanking cusps on rachidian: セ@ (0), 1(1),0 (2)
9) central cusp on rachidian: strongest (0), as strong as others (1)
10) lateral tooth: present (0), absent (1)
11) no. of ュ。セァゥョャウ@
per half row: 2 (Fig. 12) (0), '?:.7 (Fig. 13) (1)
12) no. of cusps on outer marginal: '?:.3 (0), 2 (1)
The radular characters used in this study are derived from the data shown in
Table 1 (Appendix). The apparent intraspecific variability reflected in the
table is often caused by the different microscopic techniques applied; when
possible, data based on SEM studies have been used here.
A true radula is lacking in Discotectonica and Granosolarium (Haszprunar,
1985d,e). Instead, these genera have extremely long, toothed, rod-like structures
(see Melone, 1975: pl.1 figs. 5-6; Melone & Taviani, 1985: figs. 4-10; Haszprunar,
1985d: fig. 2). The homology and function of this structure are unclear.
The total number of transverse rows of radular teeth might be a useful
taxonomic character: e.g., with 60-70 rows counted in Architectonica s.s. and only
nine in a species of Solatisonax [Boss & Merrill, 1984: 358, 361; for A. nobilis
Roding, 1798, and "Heliacus" borealis (Verrill & Smith in Verrill, 1881), respectively]; not enough data were available to use this character in the analysis.
Opercular characters
13) opercular peg: absent (Fig. 3) (0), present (Fig. 2) (1)
14) operculum shape: circular (0), oval (1)
Only two opercular characters were used. No distinction was made between
circular flat and circular high-spired (cone-shaped) opercula, because this
character varies greatly within genera [compare, e.g., the closely related
Heliacus (Teretropoma) infundibuliformi$ (Gmelin, 1791) and Heliacus (Teretropoma) mighelsi (Philippi, 1853) (see Bieler, 1985b: 104)]. Although esta-
Architectonicidae Phylogeny
219
blished in the architectonicid literature, grouping by paucispiral, "moderately
multispiral," and multispiral opercula was omitted, because such splitting was
judged to be artificial (all architectonicid opercula are originally multispiral
and differ in later whorls, as expressed by character 14, circularI oval). The oval
shape of mathildid opercula is seen here as convergence, based on a different
"ground plan" (initial "open coiling" with peg formation in Architectonicidae,
versus "normal growth" of a flat operculum in Mathildidae). To prevent the superficial resemblance of the mathildid operculum from influencing the analysis,
this character was coded as "missing data" for the outgroup.
Protoconch characters
15) protoconch size: small (0), medium (1), large (2)
16) protoconch coiling: heterostrophic (0), almost paucispiral (1)
[Coded as (1) for Pseudomalaxis and Spirolaxis. Also true for some species
of Discotectonica and Pseudotorinia (Bieler, 1985a: 241, 1985b: 92).)
17) anal keel: absent (0), developed (1)
The protoconch size in Architectonicidae has been shown to be a good
taxonomic character at the species level (Robertson, 1970; Bieler, 1984b). On the
generic level, the taxa again display different size ranges (Fig. 16), and are here
arbitrarily grouped into three size classes (character 15). With reference to heterostrophy, Haszprunar (1985e: 207) stated: "Gegania valkyrie shows only traces
of this condition (see SEM photographs in Climo 1975)." Climo's photographs
(1975: fig. 1) were taken at angles that do not emphasize the heterostrophic
condition. The shell apex of the specimen for which Haszprunar (1985e) studied
the anatomy clearly shows heterostrophy (Fig. 14), the axes of coiling being
different in protoconch and teleoconch, and the early protoconch whorls pointing
towards the base of the teleoconch.
Teleoconch characters
18) basal keel, formed by a basal field rib (Fig. 18): absent (0), present (1)
19) basal keel forming lower peripheral shell keel: no (0), yes (1)
20) infraperipheral rib forming lower peripheral shell keel (Fig. 19): no
(0), yes (1)
21) 3 narrow, well-defined outer basal field ribs (Fig. 20): absent (0),
present (1)
22) 5 basal field ribs, expanding towards umbilicus: absent (0), present (1)
23) outer basal field: sculptured (0), smooth (1)
24) concave zones flanking peripheral keel (Fig. 21): absent (0), present
:
(1)
25) peripheral ribs: sculptured (0), smooth (1)
26) mid-rib area with many secondary ribs: no (0), yes (1)
27) areas of fused ribs smooth and glossy: no (0), yes (1)
28) umbilical crenae: prominent (0), reduced (1)
29) teleoconch coiling: orthostrophic (0), planispiral (1)
Several teleoconch characters that have been used in the past to classify architectonicids have proven to be unreliable: The adult size can vary greatly
within genera (especially in Heliacus, Granosolarium and Pseudotorinia). The
general shell shape can be very misleading when not based on homologous struc-
Rudiger Bieler
220
tures ["Heliacus (Claraxis)" sensu Garrard (1977: 554) for instance, mainly based
on having "one sharp peripheral keel," included 7 Recent species from 4 genera:
Granosolarium, Pseudotorinia, Solatisonax and Architectonica).
However, shell shape based on homologous sculptural elements is a very good
SMALL
MEDIUM
LARGE
Architectonics
new taxon
Philippia
Psilaxis
Basisulcata
Heliacus
Discotecton lea
Granosolarium
Solatisonax
Pseudotorinia
Pseudomalaxis
Spirolaxis
0.75
1.20mm
FIG. 16. Protoconch size ranges of architectonicid genus-group taxa (chosen size classes
used as character 15).
taxonomic character in this family. Using the homology criteria of position, ontogenetic and fossil evidence, and special morphological qualities such as color
pattern and sculpture, a number of main spiral ribs and fields on the teleoconch
have been homologized within the Architectonicidae (Bieler, 1984b: 457, fig. 1,
1985b: 96, figs. 1-7; see Figs. 1 and 17). (In some smooth [Philippia) or highly
Architectonicidae Phylogeny
221
granulated [Heliacus (Gyriscus)] groups this "ground pattern" is only traceable in
the early whorls, and sometimes the fossil record reveals the placement of ribs
"missing" in Recent species [e.g., the Pliocene Solarium bicinctum Cantraine, 1842,
displays a weak upper peripheral rib, lacking in Recent species of Philippia s.s.;
Bieler 1985a: 236)).
SSR
セpr@
UC
UC
SSR
d SSR
セMO@
UPR
セ@
LPR
UC
UC
FIG. 17. Placement of homologous ribs in various architectonicid genera. Apertural view;
arrow shows point of attachment of next whorl, intrasubgeneric variation indicated by dotted
lines. a) Architectonica s.s., b) Philippia s.s., c) Heliacus s.s., d) Pseudomalaxis s.s.
19
FIGS. 18-21. Shape of body whorl and placement of homologous ribs in Architectonicidae.
Filled circles, from top to bottom: UPR, LPR, IPR; open circles: additional ribs formed by BF.
18. Character 18; 19. Character 20; 20. Character 21; 21. Character 24.
Other characters
Fragmentary data on additional characters are available [such as known food
requirements: Architectonica - Actiniaria (e.g., Bandel, 1984: 63), Psilaxis - Scleractinia (see Robertson et al., 1970: 60), Heliacus - Zoanthidea (e.g., Robertson,
1967: 246), Gegania - Antipatharia (see Climo, 1975: 278), but could not be utilized
in this study.
Rudiger Bieler
222
Tree Construction
Coding of multistate characters:
The multistate characters 2 and 8 (buccal mass size: large/medium/small;
number of flanking cusps on rachidian: 2+ /1/0) were coded for different runs in the
following way: hypothesis for transformation series (a) ordered (PAUP, PHYSYS): 0-1-2 or 2-1-0
0-1-2 or 0-2-1or2-1-0 or 2-0-1or1-0-2or1-2-0
(b) unordered linear (PAUP):
(c) branching (PHYSYS):
O< セ@
The multistate character 15 (protoconch size) was always coded as ordered
linear (small-medium-large or large-medium-small).
The two programs, PAUP and PHYSYS/WAGNER.S, constructed identical
cladograms. The analyses resulted in pairs of equally parsimonious trees that
were congruent except for the relative positions of Philippia and Psilaxis. The
shortest trees (length = 40) were found when the character states of characters 2
and 8 were arranged in "logical" sequence (large-medium-small and many-onenone, respectively), rather than when postulating independent derivations
(''branching") of stages 1 and 2 in these characters (length = 42, identical tree
topology). The two most parsimonious trees as illustrated in Fig. 22 will be
discussed.
DISCUSSION
The two resulting, equally parsimonious trees differ in the arrangement of Architectonica, Psilaxis and Philippia (Fig. 22a,b). The difference is caused by a
hypothetical parallelism in character 15 in one tree (Fig. 22a) and a reversal in
character 9 in the other (Fig. 22b). The former hypothesis is here preferred,
since parallel evolution in character 15 (protoconch size), a character that
displays homoplasy also in other parts of the tree, seems much more likely to
have occurred than a reversal from a specialized rachidian organization
(character 9). The following discussion is based on this cladogram (Fig. 22a).
The tree is fully resolved (no polychotomies), but the various branches (indicated
by letters in Fig. 22) are not equally well supported by the present data.
Branches A, H and K are well supported by suites of anatomical and radular
characters, indicating monophyly of Architectonicidae, Architectonica +new
taxon, and Discotectonica + Granosolarium, respectively. Branch Fis supported
by a radular character (8, reduced number of flanking cusps on rachidian) as well
as by opercular and shell characters. Branches C, E, G and L are each defined by
a number of shell characters.
Branches Band Dare defined by a single shell character each, one of which
(character 15 in B) displays parallelism elsewhere in the tree. The placement of
Heliacus is therefore problematic: besides these two characters and the synapomorphies of the family (branch A), it shares with other genera only two additional homoplasious characters (17 and 20, discussed below).
FIG. 22a,b (Opposite page). Most parsimonious cladograms of Recent architectonicld
genus-group taxa, based on data matrix shown in Table 2 (Appendix). Number to the left of
arrow =character, number to the right =character state; right arrow = gain of apomorphic
character state; left arrow = reversal; double line = homoplasy; letters = branches as referred
to in the text Length =40, consistency index = 0.80.
223
Architectonicidae Phylogeny
OI
::c
Architectonlca
II
I\)
セn@
N--N
new taxon
u' ' '
N
N
Q
Q
セ@
w
N
c...
,,
Psllaxls
IO
'
::n).l:J:m
Aセ@
Phlllppla
セ@
c
セ@
Baslsulcata
m
II
N
Hellacus
セ}@
CID
セ@
.!.!.
•!!.
セ@
Dlscotectonlca
.... U1 W _.
y Ay y
... o--
m
lU
)>
Granosolarlum
セ@
.!!.
-''
,,,,,,
-----セ。giuャn@
Solatlsonax
Pseudotorlnla
n
セウ@
II
f
u;
,...
Pseudomalaxls
II
!!!!
Splrolaxls
OUTGROUP
m
N
N
Archltectonlca
II
I\)
anセco@
''''
--ON
u;
'
new taxon
N
Q
セg@
y
II
Psilaxis
y
セ@
Phlllppla
224
Rudiger Bieler
Homoplasious characters
Seven of the 29 characters have "less-than-ideal" character state distributions in the cladogram (i.e., a consistency index <1):
Character 5 (proboscis sheath length):
The reversal to a short proboscis sheath length in branch K, comprising Discotectonica and Granosolarium, may be due to the overall modification of the alimentary tract, including the replacement of the radula by a rod-like structure,
and the loss of a buccal cavity.
Character 8 (number of flanking cusps on rachidian):
The total reduction of the flanking cusps on the rachidian is the final step in
the modification from a "taenioglossate" rachidian to a "ptenoglossate" tooth
(the first step appears in branch F), and might have occurred in parallel in Architectonica and Basisulcata, possibly due to similar adaptations as cnidarian
predators (see below).
Character 14 (operculum shape):
The change from circular- to oval-shaped opercula occurs twice, in branch F
and again, as an autapomorphy, in Discotectonica. Discotectonica and the taxa
defined by branch F comprise the species with largest shell size in the family.
Whenever architectonicids reach a critical size limit, there has to be compromise
between the optimal body volume per whorl and the ability to move a large shell
in the sand (where most members of the family spend at least their resting
periods). Large architectonicids of various genera develop a more or less sharp
peripheral shell keel that allows easier burrowing in the sand. This results in a
triangularly shaped aperture and, in tum, a compatible (i.e., oval) operculum.
Character 15 (protoconch size):
The character states "protoconch medium sized" and "protoconch large" both
show parallelism in the cladogram. This character, with its arbitrarily chosen
size classes, is probably the weakest in the analysis. The development of very
large protoconchs in several groups, however, might be explained by a directly
selective pressure on the larvae at settlement [Robertson et al. (1970: 63),
proposed an adaptive advantage to large protoconchs in Psilaxis that precludes
their being swallowed by the polyps of their host corals].
Character 17 (anal keel):
An anal keel [as defined by Robertson, 1964: 3, fig. 10) on the larval shell is
present in all Recent Philippia and Psilaxis species (but lacking in some fossil
members attributed to these genera) and is more or less well-developed in a few
species of Architectonica, of the new taxon, and of Heliacus. This, the occurrence
of such a keel in the Paleocene architectonicid genus Dinaxis (see Bieler, 1985a:
239, pl. 2 fig. 10), and a similar (homologous?) structure in Gegania valkyrie (Fig.
14), indicate that this feature is not very reliable as a taxonomic character and is
probably older than indicated by the cladogram.
Character 20 (lower peripheral shell keel formed by IPR):
The infraperipheral rib (IPR) serves as the lower peripheral shell keel in
Philippia and Heliacus. Both groups have relatively high-spired shells with
rounded whorls, that facilitate movement on and between polyps in cnidarian
colonies. Similar shell shape, and resulting identical placement of the IPR, is
here interpreted as convergent.
Character 21 (3 narrow outer basal field ribs):
Architectonicidae Phylogeny
225
The parallel occurrence of this character in branch C (leading to Pseudotorinia, Pseudomalaxis and Spirolaxis) and as an autapomorphy of Granosolarium is
most difficult to interpret. In its complexity (number, position and sharp definition of the ribs) it was initially thought to be a good taxonomic character uniting
Pseudotorinia, Pseudomalaxis and Granosolarium. The entirely different alimentary tract (Haszprunar, 1985e) however, groups Granosolarium with Discotectonica (branch K).
Placement of extinct taxa
A number of architectonicid genus-group taxa are known only from fossils.
Based on proto- and teleoconch characters alone, three of them can be placed in
the present cladogram (Fig. 23):
Stellaxis Dall, 1892 (described from the Eocene): Members of this genus have a
more or less smooth, cone-shaped teleoconch (Bieler, 1985a: 240, pl. 2 fig. 11, pl. 3
fig. 15), that is very similar to the organization found in Architectonica, Psilaxis
and Basisulcata. The small- to medium-sized protoconch has no anal keel.
Nipteraxis Cossmann, 1915 (described from the Eocene): This group is very
close to Heliacus in shell characters, except for the organization of the peripheral keel [the upper peripheral rib (UPR) is much heavier and the infraperipheral rib (IPR) is much less developed; Bieler, 1985b: 94, 96 fig.2, pl.1 fig. 2].
Climacopoma Fischer, 1885 (described from the Eocene): The characters
comprised by branches B and E in Fig. 22 group this genus with Discotectonica,
Granosolarium and Solatisonax (Bieler 1985a: 244, pl. 4 figs. 16-18).
Two monotypic genera, Eosolarium Chavan, 1947 [described from the Upper
Cretaceous (Campanium)], and Dinaxis Dall in Aldrich, 1895 (described from the
Paleocene) could not be placed in the tree. Eosolarium lacks sufficient information on protoconch characters and teleoconch shape due to the poor condition of
the type material (Bieler, 1985b: 106, pl. 4 fig. 15). The unique shape and
sculpture of the juvenile type specimen of Dinaxis did not allow classification
with any of the Recent taxa (Bieler, 1985a: 239, pl. 2 fig. 10).
Evidence from the fossil record
Merrill (1970: 273, fig.1) proposed a tree including nine architectonicid genusgroup taxa here considered valid, arranged by similarity of the taxa and, based
on the fossil record, by their appearance in time (Fig. 24). The branching sequence
of this tree is remarkably similar to that of the cladogram (Fig. 22). Merrill's
time scale could, with two exceptions, be directly applied to the cladogram (Discotectonica, as Acutitectonica, was grouped with Architectonica and Philippia
by Merrill, because it was believed to have a similar radula in addition to the
rod-like structure; Solatisonax appears relatively late in the time scale because
some species now included were then thought to belong to Heliacus).
Thus, the overall branching pattern of the cladogram is supported by fossil
evidence, with the groups having sculptured shells, circular opercula and
"taenioglossate-like" radulae in the Recent forms evolving earlier than those
with smooth shells, oval opercula and "ptenoglossate-like" radulae.
Evidence from ontogeny
In some cases the direction of character state change can be directly inferred
Rudiger Bieler
226
...>
Architectonic a
new taxon
n
)I
'i\rchitectonicinae"
n
I»
Psilaxis
Philippia
--------
)f
"Philippiinae"
>
"Heliacinae''
t Stellaxis
Basisulcata
- - - - - - - - - - · tNipteraxis
Heliacus
Discotectonica
Granosolarium
m
-----· t Climacopoma
Solatlsonax
..
Pseudotorlnia
n
Pseudomalaxis
r-
セ@
H
"Pseudomalaxlnae"
Spirolaxls
ii
)C
enFIG. 23. Modified cladogram of Architectonicidae, including fossil genus-group taxa.
Dotted lines indicate tentative placement of extinct taxa; thick branches mark wellsupported clades; subfamily names of authors in quotation marks.
Architectonicidae Phylogeny
227
from ontogenetic evidence: all oval-shaped opercula in this family are circular in
their early whorls (character 14), and the direction from sculptured to smooth
states in teleoconch characters 23, 25 and 27 can be demonstrated in juvenile shells
of Architectonica and Philippia, where traces of the presumed ancestral character
states can often be seen on the first teleoconch whorl.
セ@
セ@
(,)
(,)
--
'E
-:c0
(,)
Cl>
...
(,)
"'
セ@
a.
セ@
:c
(I)
';;(
セ@
セ@
セ@
(,)
c
Cl>
(I)
(,)
c
0
:s
=
0 a;
0
en
:c
:s
Q
::::!
·;;c
セ@
:c
0...
a.
:s
Q)
"'
Q
(I)
0
'O
Q)
(I)
セ@
セ@
E
0
'O
(,)
.!
セ@
セ@
-
(I)
0
w
I(
:5
I(
セ@
(I)
0
.!
セ@
·r:0
(I)
セ@
"'
セ@
"'
セ@
I
1,
RECENT
''
MIOCENE
I
I
/
セG@
/
'-......
セ@
/
セ@
EOCENE
PALEOCENE
BGMNセ@
セ@
セ@
CRETACEOUS
JURASSIC
TRIASSIC
"'-...
FIG. 24. Dendrogram of the Architectonicoidea, arranged by sequence of appearance in
time (based on fossil evidence). Vertical scale compressed and only periods significant to Architectonicoidea evolution shown within each era. Time dimensions not proportional
[Revised after Merrill (1970: pl. 42): manuscript name, non-architectonicid genus and subgenus
Heliacus (Gyriscus) omitted; for pairs of nominal taxa here regarded as synonyms (e.g., Spirolaxis-Paurodiscus, Pseudotorinia-Awarua) only earliest placement used; distinction in generic
and subgeneric levels omitted; arrangement of taxa modified to facilitate easier comparison
with cladogram in Fig. 22).
Comparison with earlier ·classifications
It was in a way unfortunate that of all the architectonicid genera, Architectonica has the most derived radula in the family. Because Architectonica is the
type genus of this family, the Architectonicidae were subsequently considered to
be ptenoglossate by various authors. Fretter & Graham (1982: 364) for instance,
see a subgroup within the "Heterogastropoda": Architectonicidae, Epitonioidea
228
Rudiger Bieler
and Eulimoidea" ... whose radular teeth are numerous and homodont." In order to
explain the occurrence of radulae with fewer teeth per row in the family, a
reduction had to be postulated: Climo (1975: 281) interpreted the reduction as a
reflection of the smaller adult size, and Golikov & Starobogatov (1975: 222)
placed the Architectonicidae and Mathildidae in a suborder Gymnoglossa Gray,
1853, as members who "retain their radula (which shows a gradually decreasing
number of teeth)."
The ptenoglossate-like radula in Architectonica is clearly a derived condition
and most likely evolved as an adaptation to its specialized feeding on cnidarians. Similarity with the radulae of Epitoniidae and Janthinidae can only I>e interpreted as convergence (see also Thiele, 1928: 73). The "typical" architectonicid radula is "five-toothed taenioglossate-like" with two pairs of almost
identical marginals as illustrated in Fig. 12.
Due to the high degree of variability displayed in shell, opercular, and especially radular characters in this family, the Architectonicidae have been
variously split into two families [Solariidae and Toriniidae: Troschel (1875),
Schepman (1909); Architectonicidae and Heliacidae: Cotton & Godfrey (1933),
Rippingale & McMichael (1961); Architectonicidae and Mangonuidae: Iredale
(1936)), and some suggested further splitting: "The Architectonicidae and Toriniidae, though, are rather heterogeneous as to their radulae and may have to be
divided into a number of smaller families" (Golikov & Starobogatov, 1975: 223)."
Others (e.g., Marshall, 1887; Powell, 1965; Abbott, 1974) used two or more subfamilial divisions.
At least the following eight family-group names have been proposed:
for "Architectonica-group":
- Architectonicidae Gray, 1850 (as "Architectomidae," based on Architectonica
Roding, 1798).
- Solariidae Chenu, 1859 [based on Solarium Lamarck, 1799 (= Architectonica)].
for "Philippia-group":
- Philippiinae Boss, 1982 (based on Philippia Gray, 1847).
for "Heliacus-group":
- Toriniidae Troschel, 1875 [as family, ''Toriniacea," based on Torinia Gray,
1842 (= Heliacus Orbigny, 1842)).
-Teretropomidae Rochebrune, 1881 (based on Teretropoma Rochebrune, 1881 [=
Heliacus (Teretropoma)], as a family of land snails}.
- Heliacidae Cotton & Godfrey, 1933 (based on Heliacus).
for "Pseudomalaxis-group":
- Mangonuidae Iredale, 1936 [based on Mangonuia Mestayer, 1930 (=Pseudomalaxis Fischer, 1885)].
- Pseudomalaxinae Garrard, 1977 (based on Pseudomalaxis).
Three recent revisions used subfamilies as categories in the Architectonicidae
(Table 3, Appendix). Garrard (1977), in his revision of Australian architectonicids,
used mainly shell shape and opercula as taxonomic characters; Boss & Merrill
(1984) exclusively used characters of radulae and jaws; and Melone & Taviani
(1985) used shell, opercular and radular characters. None of the three taxonomic
arrangements agrees with the data presented here. For instance, the "rod-like
structure," a highly derived character shared by species of the genera Discotectonica and Granosolarium (indicated by"*" in Table 3), indicates a very close re-
Architectonicidae Phylogeny
229
lationship of the two. In Garrard's system the two genera were separately
placed in subfamilies Architectonicinae and Heliacinae; in Boss & Merrill's classification at least one of them (Discotectonica; Granosolarium was not covered) is
placed in the Philippiinae. The results of this analysis (Fig. 23) do not support
splitting of the family into previously defined subfamilies. Three wellsupported clades (and Heliacus) can be distinguished within the family, and all
are incongruent with the established subdivisions. Pairs of taxa that form monophyletic groups (Architectonica/new taxon, Philippia/ Psilaxis, Pseudomalaxis/
Spirolaxis) can be treated as genera (Architectonica, Philippia, Pseudomalaxis,
respectively), but the subfamilies of authors (Table 3) are largely polyphyletic
assemblages (the "Heliacinae" are usually defined by shared ancestral character
states).
CONCLUSIONS
Two basic problems in earlier classifications of the Architectonicidae have
been encountered: (1) the misinterpretation of the type genus as being "typical"
for the family, and (2) the fact that no distinction was made between features
that can be used to describe a taxon and characters that define monophyletic units
(i.e., syn- and autapomorphies).
The family Architectonicidae is a monophyletic group that is defined by a
number of synapomorphies in anatomical, radular and opercular characters.
Splitting the group into several families is not justified.
Architectonicidae and Mathildidae form the superfamily Architectonicoidea, which is defined by a suite of anatomical synapomorphies and the
common possession of a heterostrophic protoconch (a symplesiomorphy).
Opercular and radular characters, however, are not synapomorphic, with the
"five-toothed-taenioglossate-like" radulae of Architectonicidae and Mathildidae having evolved independently by parallel reduction of non-homologous
teeth.
The most parsimonious cladogram (the one that involves the fewest number of
"ad hoc hypotheses" of homoplasy) of the architectonicid genus-group taxa
based on anatomical, radular, opercular and shell characters shows three
distinct, well-supported clades (Architectonica + new taxon + Philippia +
Psilaxis + Basisulcata; Discotectonica + Granosolarium + Solatisonax; Pseudotorinia + Pseudomalaxis + Spirolaxis). The position of Heliacus in the tree is
weakly supported (Heliacus is mainly defined by unique derived characters and
retained symplesiomorphies). The relative branching sequence in the cladogram
is supported by fossil and ontogenetic evidence. Three fossil taxa (Stellaxis, Nipteraxis, Climacopoma) can be added to the tree, based on shell characters alone.
Architectonicids with smooth shells, oval-shaped opercula and "ptenoglossatelike" radulae apparently evolved later than those with sculptured shells,
circular opercula and "taenioglossate-like" radulae. The use of the
"ptenoglossate-like" radula as a character to combine Architectonicidae with
mesogastropod families such as Epitoniidae and Janthinidae is not justified; the
similar radular type must be explained by convergence.
In order to derive a classification from the cladogram it is necessary to assign
absolute rank to the monophyletic groups inferred: the 12 architectonicid taxa
Rudiger Bieler
230
that have been tenned "genus-group taxa" throughout the study can either all be
interpreted as separate genera, or, several of them if they form monophyletic
groups, can be treated as subgenera of a single genus (e.g., Architectonica,
Philippia, Pseudomalaxis). The published subdivisions of the Architectonicidae
into subfamilies are not congruent with the groups found in this study; they were
not based on monophyly. For a small family such as this, the need for or
advantage in using and naming subfamily units is doubtful.
ACKNOWLEDGEMENTS
Most data on Architectonicidae used in this study were gathered for a Ph.D. thesis (University of
Hamburg, Federal Republic of Germany) under Prof. Dr. Otto Kraus, whose advice and support are
gratefully acknowledged. That part of the work was funded by a stipend of the "Studienstiftung des
Deutschen Volkes" Bonn, Federal Republic of Germany. Additional studies on Architectoniddae and
Mathildidae were made during Smithsonian Postdoctoral Fellowships at the National Museum of
Natural History, Washington, and at the Smithsonian Marine Station at Link Port, Florida, U.S.A.
Material from 59 collections was studied (see acknowledgements in my publications cited). Space
forbids me to thank again all the helpful individuals and institutions who made this study possible. I
am indebted to Dr. Gerhard Haszprunar (University of Vienna, Austria) for supplying serial sections
of mathildids, to Ms. Paula M. Mikkelsen (Indian River Coastal Zone Museum, Ft. Pierce), Dr. James
F. McLean (Los Angeles County Museum), Mr. Bruce Marshall (National Museum of New Zealand,
Wellington), and Dr. John D. Taylor [British Museum (Natural History), London] for supplying additional alcohol-preserved architectonicid specimens, to Ms. Elana Benamy (Academy of Natural
Sciences of Philadelphia) and Mr. Ian Loch (Australian Museum, Sydney) for the loan of additional
architectonicid and mathildid specimens, to Dr. Anders Waren (Swedish Museum of Natural History,
Stockholm) for providing previously unpublished SEM photomicrographs of mathildid radulae, and
to Mr. Gary Rosenberg (Harvard University, Cambridge) for helping with literature searches. The
PAUP program (version 2.40) was run on an IBM-AT personal computer, PHYSYS was run on an
UNIVAC mainframe computer with the assistance of Dr. Mary F. Mickevich (USNM). Scanning
electron micrographs.of gold/palladium-coated specimens were taken on a Zeiss NOVASCAN-30
machine at the Smithsonian Marine Station at Link Port, with the assistance of Ms. Julianne Piraino.
I thank Dr. Kenneth J. Boss (Museum of Comparative Zoology, Harvard University, Cambridge), Dr.
Jonathan A. Coddington (USNM), Dr. M.G. "Jerry" Harasewych (USNM) and Ms. Paula M. Mikkelsen
(IRCZM) for critical comments on this manuscript.
LITERATURE CITED
ABBOTT, R.T. 1974. American Seashells. The Marine Mollusca of the Atlantic and Pacific Coasts of
North America [2nd ed.]. Van Nostrand Reinhold Co., New York. 663 pp.
ALDRICH, T.H. 1895. Descriptions of two new Eocene Solariidae from Alabama. The Nautilus, 9: 1-
2
BANDEL, I<. 1984. The radulae of Caribbean and other Mesogastropoda and Neogastropoda. Zoologische Verhandelingen, 214: 1-188.
BAYER, C. 1940. Catalogue of the Solariidae in the Rijksmuseum van Natuurlijke Historie. I.
Solarium s.s. Zoologische Mededelingen, 22: 223-256.
BIELER, R 1984a. Die Gattungen der Architectonicidae (Gastropoda: "Heterogastropoda"). Allgemeines und Tell 1: Pseudomalaxis. Archi'o ftlr Molluskenkunde, 115: 53-103.
BIELER, R. 1984b. Morphometrische Analyse der Architectonica maxima-Gruppe im lndo-Pazifik
(Mollusca: Gastropoda: Architectoniddae). Verhandlungen des naturwissenschaftlichen Vereins
in Hamburg, (NF) 27: 453-492.
BIELER, R. 1985a. Die Gattungen der Architectonicidae (Gastropoda: "Heterogastropoda''). Tell 2:
Architectonica, Philippia, Dinaxis, Stellaxis, Discotectonica, Solatisonax, Climacopoma, Granosolarium. Archiv ftlr Molluskenkunde, 115: 231-265.
BIELER, R. 1985b. Die Gattungen der Architectonicidae (Gastropoda: Allogastropoda). Tell 3: Pseudotorinia, Nipteraxis, Heliacus, Eosolarium. Archiv ftlr Molluskenkunde, 116: 89-117.
BIELER, R. 1986. Revision of genera and lndo-Pacific species in the family Architectonicidae
(Abstract, American Malacological Union 51st Annual Meeting). American Malacological
Bulletin, 4: 108-109.
BIELER, R. 1987. Die Gattungen der Architectonicidae (Gastropoda: Allogastropoda). Tell 4:
Architectonicidae Phylogeny
231
Heliacus (Pyrgoheliacus) n.subgen. und Architectonica (Adelphotectonica) n.subgen. Archi'o far
Molluskenkunde, 117: 203-215.
BOSS, K.J. 1982. Mollusca [and] Oassification of Mollusca. In: Parker, S.P. (Ed.), Synopsis and Classification of LiTJing Organisms. McGraw-Hill, New York; Vol.1: 945-1166, Vol.2: 1092-1096.
BOSS, K.J. &: MERRILL, AS. 1984. Radular configuration and the taxonomic hierarchy in the Architectonicidae (Gastropoda). Occasional Papers on Mollusks, 4(66): 349-411.
BOUVIER, E.L 1886a. Contributions a l'etude des Prosobranches ptenoglosses. Bulletin de la Sociitt
malacologique de France, 3: 77-130.
BOUVIER, E.L. 1886b. Observations anatomiques relatives aux Solarides et aux Janthinides.
Bulletin de la Sociitt philomatique de Paris, (7)10: 151-156.
BOUVIER, E.L 1887. Systeme nerveux, morphologie generale et classification des Gasteropodes prosobranches. Annales des Sciences naturelles (Zoologie), (7)3: 1-510.
CERNOHORSKY, W.O. 1972. Marine Shells of the Pacific, II. Pacific Publ., Sydney. 411 pp.
CERNOHORSKY, W.O. 1978. Tropical Pacific Marine Shells. Pacific Publ., Sydney. 352 pp.
CHAVAN, A. 1947. La faune campanienne du Mont des Oliviers d'apres les materiaux VignaJ-Masse.
Journal de Conchyliologie, 87: 125-197.
CHENU, J.C. 1859. Manuel de Conchyliologie et de Paltontologie conchyliologique, 1: vii+508 pp. V.
Masson, Paris.
CLIMO, F.M. 1975. The anatomy of Gegania ?Jalkyrie Powell (Mollusca: Heterogastropoda: Mathildidae) with notes on other heterogastropods. Journal of the Royal Society of New Zealand, 5: 275288.
COSSMANN, M. 1896. Essais de Paltoconchologie comparie, 2: 179 pp. Paris.
COSSMANN, M. 1915. Essais de Paltoconchologie comparie, 10: 292 pp. Paris.
COITON, B.C. &: GODFREY, F.K. 1933. South Australian shells (including descriptions of new genus
and species), Part VII. South Australian Naturalist, 14: 72-108.
DALL, W.H. 1892. Contributions to the Tertiary fauna of Florida, with especial reference to the
Miocene Silex-beds of Tampa and the Pliocene beds of the Caloosahatchie River, II. - Streptodont
and other gastropods, concluded. Transactions of the Wagner Free Institute of Science, 3: 201-458.
EAMES, F.E. 1952. A contribution to the study of the Eocene in Western Pakistan and Western India.
C. The description of the Scaphopoda and Gastropoda from standard sections in the Rakhi Nala
and Zinda Pir areas of the western Punjab and in the Kohat District. Philosophical Transactions
of the Royal Society of London, (B)236(631): 1-168.
FARRIS, J.S. 1970. Methods for computing Wagner trees. Systematic Zoology, 19: 83-92.
FARRIS, J.S. 1975. Formal definitions of paraphyly and polyphyly. Systematic Zoology, 23(4): 548554.
FINLAY, H.J. 1927. A further commentary on New Zealand molluscan systematics. Transactions and
Proceedings of the New Zealand Institute, 57: 320-485.
FISCHER, P. 1885. Manuel de Conchyliologie et de PaUontologie conchyliologique ou Histoire
naturelle des Mollusques TJivants et fossiles (suivi d'un appendice sur les Brachiopodes par D.P.
Oehlert). (8): 689-784 (1885). F. Savy, Paris.
FITCH, W.M. 1971. Toward defining the course of evolution: minimum change for a specified tree
topology. Systematic Zoology, 20: 406-416.
FRETIER, V. 1948. The structure and life history of some minute prosobranchs of rock pools: Skeneopsis planorbis (Fabricius), Omalogyra atomus (Philippi), Rissoella diaphana (Alder) and Rissoella
opalina (Jeffreys). Journal of the Marine Biological Association of the United Kingdom, 27: 597632.
FRETTER, V. &: GRAHAM, A. 1949. The structure and mode of life of the Pyramidellidae, parasitic
opisthobranchs. Journal of the Marine Biological Association of the United Kingdom, 28: 493-532.
FRETIER, V. &: GRAHAM, A. 1982. The prosobranch molluscs of Britain and Denmark. Part 7 'Heterogastropoda' (Cerithiopsacea, Triforacea, Epitoniacea, Eulimacea). Journal of Molluscan
Studies, Supplement 11: 363-434.
FRETTER, V., GRAHAM, A. &: ANDREWS, E.B. 1986. The prosobranch molluscs of Britain and
Denmark. Part 9 - Pyramidellacea. Journal of Molluscan Studies, Supplement 16: 557-649.
GARRARD, T.A. 1961. Mollusca collected by M.V. "Challenge" off the east coast of Australia.
Journal of the Malacological Society of Australia, 1(5): 2-37.
GARRARD, T.A. 1977. A revision of Australian Architectonicidae (Gastropoda: Mollusca). Records
of the Australian Museum, 31: 506-584.
GHISELIN, M.T., DEGENS, E.T., SPENCER, D.W. &: PARKER, R.H. 1967. A phylogenetic survey of
molluscan shell matrix proteins. Breviora, 262: 1-35.
GIUSTI, F. &: SELMI, M.G. 1982. The atypical sperm in the prosobranch molluscs. (Proceedings of the
7th International Malacological Congress), Malacologia, 22: 171-181.
GOLIKOV, A.N. &: STAROBOGATOV, Y.I. 1975. Systematics of prosobranch gastropods. Malacologia, 15: 185-232.
GOSLINER, T.M. 1981. Origins and relationships of primitive members of the Opisthobranchia
232
Rudiger Bieler
(Mollusca: Gastropoda). Biological Journal of the Linnean Society of London, 16: 197-225.
GRAY, J.E. 1842. Synopsis of the Contents of the British Museum [44.ed.]. Mollusca: 48-92. British
Museum. London.
GRAY, J.E. 1847. List of genera of Recent Mollusca, their synonyma and types. Proceedings of the Zoological Society of London, 15: 129-219.
GRAY, J.E. 1850. [Explanation of plates and list of genera]. 124 pp. In: Gray, M.E., Figures of Mollusams Animals, 4: iv+219 pp. London.
GRAY, J.E. 1853. On the divisions of ctenobrancheous gasteropodous Mollusca into larger groups and
families. Proceedings of the Zoological Society of London, 1853: 32-44.
HABE, T. 1943. On the radulae of Japanese marine gastropods (1). Venus, 13: 68-76.
HABE, T. 1958. On the radulae of Japanese marine gastropods (4). Venus, 20: 43-60.
HABE, T. 1961. Coloured lllustrations of the Shells of Japan (II). Hoikusha, Osaka. ix+183+42 pp.
HABE, T. & KOSUGE, S. 1966. Shells of the World in Colour. Volll. The Tropical Pacific. Hoikusha,
Osaka. vii+193 pp.
HASZPRUNAR, G. 1985a. The fine morphology of the osphradial sense organs of the Mollusca. I.
Gastropoda, Prosobranchia. Philosophical Transactions of the Royal Society of London, B307:
457-496.
HASZPRUNAR, G. 1985b. The fine morphology of the osphradial sense organs of the Mollusca. II.
Allogastropoda (Architectonicidae, Pyramidellidae). Philosophical Transactions of the Royal
Society of London, B307: 497-505.
HASZPRUNAR, G. 1985c. The Heterobranchia - a new concept of the phylogeny of the higher Gastropoda. Zeitschrift filr ZDologische Systematik und Eoolutions-Forschung, 23: 15-37.
HASZPRUNAR, G. 1985d. Zur Anatomie und systematischen Stellung der Architectonicidae
(Mollusca, Allogastropoda). Zoologica Scripta, 14: 25-43.
HASZPRUNAR, G. 1985e. On the anatomy and systematic position of the Mathildidae (Mollusca,
Allogastropoda). Zoologica Scripta, 14: 201-213.
HASZPRUNAR, G. 1988. A preliminary phylogenetic analysis of the streptoneurous gastropods. In:
Ponder, W.F. (Ed.) Prosobranch phylogeny: Proceedings of a Symposium held at the 9th International Malacological Congress, Edinburgh, 1986. Malacological Ret1iew, Supplement, 4: 7-16.
(1his volume).
HEALY, J.M. 1982. Ultrastructure of spermiogenesis of Philippia (Psilaxis) orytropis, with special
reference to the taxonomic position of the Architectonicldae (Gastropoda). Zoomorphology, 101:
197-214.
HEDLEY, C. 1903. Scientific results of the trawling expedition of H.M.C.S. "Thetis" off the coast of
New South Wales in February and March 1896. Mollusca II, Scaphopoda and Gastropoda.
Memoirs of the Australian Museum, 4: 327-402.
HENDY, M.D. & PENNY, D. 1982. Branch and bound algorithms to determine minimal evolutionary
trees. Mathematical Biosciences, 59: 277-290.
HENNIG, W. 1950. Grundzage einer Theorie thr phylogenetischen Systematik. Deutsches Entomologisches Institut, Bonn-Friedrichshagen/ Deutscher Zentralverlag, Berlin. il+370 pp.
HENNIG, W. 1966. Phylogenetic Systematics. University of Illinois Press, Urbana. 263 pp.
IREDALE, T. 1929. Mollusca from the continental shelf of Eastern Australia. No.2. Records of the
Australian Museum, 17: 157-189.
IREDALE, T. 1931. Australian molluscan notes, No.1. Records of the Australian Museum, 18: 201-235.
IREDALE, T. 1936. Australian molluscan notes, No.2. Records of the Australian Museum, 19: 267-340.
IREDALE, T. 1957. An exciting find. Proceedings of the Royal Society of New South Wales, 1955-56:
124-125.
JAUME, M.L. & BORRO, P. 1946. Novedades en Moluscos marinos Cubanos. Ret1ista de la Sociedad
malacol6gica Carlos de la Torre, 4(1): 13-22.
JOUSSEAUME, F. 1888. Description des mollusques recueillis par M. le Dr. Faurot dans la Mer Rouge
et le Golfe d'Aden. Mbnoires th la Socilti z.oologique de France, 1: 165-223.
KILBURN, R.N. & RIPPEY, E. 1982. Seashells of Southern Africa. Macmillan South Africa. Johannesburg. xi+249 pp.
KLUGE, A.G. & FARRIS, J.S. 1969. Quantitative phyletics and the evolution of anurans. Systematic
Zoology, 18: 1-32.
KNIGHT, J.B. 1931. The gastropods of the St. Louis, Missouri, Pennsylvanian Outlier: Aclisina and
Streptacis. Journal of Paleontology, 5: 1-15.
KOROBKOV, I.A. 1955. Sprat1ochnik i metodicheskoe posobie po tretichnym mollyuskam. Bryukhonogie. Gostoptekhizdat, Leningradskoe otd. Leningrad. 795 pp.
KOSUGE, S. 1966. The family Triphoridae and its systematic position. Malacologia, 4: 297-324.
KURODA, T. & HABE, T. 1954. Notes on three remarkable species of Japanese gastropods. Venus, 18:
79-84.
LAMARCK, J.B.P.A. DE MONET DE. 1799. Prodrome d'une nouvelle classification des Coquilles, comprenant une redaction appropriee des caracteres genertques, et l'etabllssement d'un grand nombre
Architectonicidae Phylogeny
.
.
233
de genres nouveaux. Memoires de la Socitte d'Histoire Naturelle de Paris, 1: 63-91.
MacDONALD, J.D. 1860. Further observations on the metamorphosis of Gasteropoda, and the affinities of certain genera, with an attempted natural distribution of the principal families of the
order. Transactions of the Linnean Society of London, 23: 69-81.
MARSHALL, B. 1983. A revision of the Recent Triphoridae of Southern Australia. Records of the
Australian Museum, Supplement, 2: 1-119.
MARSHALL, W.B. 1887. Monograph of the family Solariidae. In: Tryon, G.W., Manual of Conchology; Structural and Systematic, with Illustrations of the Species, 9: 3-32. Philadelphia.
MARWICK, J. 1931. The Tertiary Mollusca of the Gisbome District. New Zealand Department of
Scientific and Industrial Research, Geological SuTVey Branch, Palaeontological Bulletin, 13: v, 1177.
MELONE, G. 1974. Note su alcuni Architectonicidae (Gastropoda, Prosobranchia) del Mediterraneo.
Quaderni della Civica Stazione Idrobiologica di Milano, 5: 23-38.
MELONE, G. 1975. Considerazioni sistematiche su un Architectonicide giapponese: Acutitectonica acutissima (G.B. Sowerby, 1914) (Gastropoda, Prosobranchia). Conchiglie, 11: 165-174.
MELONE, G. & TAVIANI, M. 1985 ["1984"]. Revisione delle Architectonicidae del Mediterraneo.
(Atti del Simposio, Bologna 24-26.IX.1982), Lavori della Societa Malacologica ltaliana, 21: 149192.
MERRILL, A.S. 1970. The Family Architectonicidae (Gastropoda: Mollusca) in the Western and
Eastern Atlantic. Ph.D. thesis, University of Delaware. 338 pp. [University Microfilms, Inc.
Ann Arbor, Michigan, No.71-6444).
MESTAYER, M.K. 1930. Notes on New Zealand Mollusca. No.5. Transactions and Proceedings of the
New Zealand Institute, 61: 144-146.
MICKEVICH, M.F. 1978. Taxonomic congruence. Systematic Zoology, 27: 143-158.
MICKEVICH, M.F. 1982. Transformation series analysis. Systematic Zoology, 31: 461-478.
MICKEVICH, M.F. & FARRIS, J.S. 1984. PHYSYS (Documentation and mainframe FORTRAN
program package).
MOORE, D.R. 1966. The Cyclostremellidae, a new family of prosobranch mollusks. Bulletin of
Marine Science, 16: 480-484.
MONTEROSATO, T.A. 1913. Note on the genus Pseudomalaxis, Fischer, and descriptions of a new
species and sub-genus. Proceedings of the Malacological Society of London, 10: 362-363.
MOQUIN-TANDON, A. 1848. Observations sur les machoires des Helices de la France. Memoires de
l'Academie Royale des Sciences de Toulouse, (3)4: 371-381.
ORBIGNY, A. d'. 1841-1846. Mollusques. In: Ramon de la Sagra, Histoire Physique, Politique et
Naturelle de l'fte de Cuba. 1: 1-208 (1841), 209-264 (1842), 2: 1-112 (1842), 113-380 ["1846" (1853?)].
PLATNICK, N.I. 1987. An empirical comparison of microcomputer parsimony programs. Cladistics, 3:
121-144.
PONDER, W.F. 1985. A review of the genera of the Rissoidae (Mollusca: Mesogastropoda:
Rissoacea). Records of the Australian Museum, Supplement 4: 1-221.
PONDER, W.F. & WAREN, A. 1988. Oassification of the Caenogastropoda and Heterostropha - A
list of the family-group names and higher taxa. In: Ponder, W.F. (Ed.) Prosobranch phylogeny:
Proceedings of a Symposium held at the 9th International Malacological Congress, Edinburgh,
1986. Malacological Review, Supplement, 4: 288-328. (This volume).
POWELL, A.W.B. 1965. New Zealand molluscan systematics with descriptions of new species: Part 5.
Records of the Auckland Institute and Museum, 6: 161-168.
POWELL, A.W.B. 1979. New Zealand Mollusca; marine, land and freshwater shells. Collins,
Auckland. xiv+500 pp.
REHDER, H.A. 1935. New Caribbean marine shells. The Nautilus, 48: 127-130.
RIPPINGALE, O.H. & McMICHAEL, D.F. 1961. Queensland and Great Barrier Reef shells.
Jacaranda Press, Brisbane. Index+210 pp.
RISBEC, J. 1955. Considerations sur l'anatomie comparee et classification des gasteropodes prosobranches. Journal de Conchyliologie, 95: 45-82.
ROBERTSON, R. 1964. Dispersal and wastage of larval Philippia krebsii (Gastropoda: Architectonicidae) in the North Atlantic. Proceedings of the Academy of Natural Sciences of Philadelphia,
116(1): 1-27.
ROBERTSON, R. 1967. Heliacus (Gastropoda: Architectoniddae) symbiotic with Zoanthiniaria
(Coelenterata). Science, 156(3772): 246-248.
ROBERTSON, R. 1970. Systematics of Indo-Pacific Philippia (Psilaxis), architectonicid gastropods
with eggs and young in the umbilicus. Pacific Science, 24: 66-83.
ROBERTSON, R. 1973. Cyclostremella: a planispiral pyramidellid. The Nautilus, 87: 88.
ROBERTSON, R. 1974 ["1973"]. The biology of the Architectonicidae, gastropods combining prosobranch and opisthobranch traits. (Proceedings of the 4th European Malacological Congress),
Malacologia, 14: 215-220.
ROBERTSON, R. 1985. Four characters and the higher category systematics of gastropods. American
234
Rudiger Bieler
Malacological Bulletin, Special Edition no. 1: 1-22.
ROBERTSON, R., SCHELTEMA, R.S. & ADAMS, F.W. 1970. The feeding, larval dispersal, and
metamorphosis of Philippia {Gastropoda: Architectonicidae). Pacific Science, 24: 55-65.
ROCHEBRUNE, A.T. DE. 1881. Sur un type nouveau de la famille des Cyclostomaceae. Bulletin de
la Socittt philomathique de Paris, (7)5: 108-115.
[RODING, P.F.] 1798. Museum Boltenianum sit1e catalogus cimeliorum e tn'bus regnis naturae (...}pars
secunda continens conchylia si'f1e testacea univalvia, bivalvia & multivalvia. J.C. Trapp,
Hamburg. viii+199 pp.
SACCO, F. 1892. I molluschi dei terreni terziarii del Piemonte e della Liguria, 12 (Pyramidellidae
(fine), Ringiculidae, Solariidae e Scalariidae]. C.Clausen, Torino. 86 pp.
SALVINl-PLAWEN, L. von & HASZPRUNAR, G. 1987. The Vetigastropoda and the systematics of
streptoneurous Gastropoda (Mollusca). Journal of Zoology, London, 211: 747-710.
SARS, G.O. 1878. Bidrag til kundskaben om Norges arktiske Fauna. I. Mollusca regionis arcticae
norvegiae. Oversigt over de i Norges arktiske Region forekommende Bleddyr. Christiania.
xiil+466 pp.
SCHEPMAN, M.M. 1909. The Prosobranchia of the Siboga Expedition. Part II. Taenioglossa and
Ptenoglossa. Resultats Siboga-Expeditie, Monograph 49(1b): 109-231.
SHUTO, T. 1969. Neogene gastropods from Panay Island, the Philippines. Contributions to the
geology and palaeontology of Southeast Asia, LXVIII. Memoirs of the Faculty of Science, Kyushu
University, Series D, Geology, 19: 1-250.
SWOFFORD, D.L 1985. PAUP- Phylogenetic Analysis Using Parsimony, Version 2.4 {Documentation
and IBM PC-compatible FORTRAN program package). Illinois Natural History Survey.
TAYLOR, D.W. & SOHL, N.F. 1962. An outline of gastropod classification. Malacologia, 1(1): 7-32.
THIELE, J. 1921. Zur Systematik der Mollusken. Archiv fiJ.r Molluskenkunde, 53:140-163.
THIELE, J. 1925. Gastropoda der Deutschen Tiefsee-Expedition. II.Tell. Wissenschaftliche Ergebnisse der Deutschen Tie/see-Expedition au/ dem Damp/er "Valdivia" 1898-1899, 17: 35-282 (1348).
THIELE, J. 1928. Ober ptenoglosse Schnecken. Zeitschrift fiJ.r wissenschaftliche Zoologie, 132: 73-94.
THIELE, J. 1929. Handbuch der systematischen Weichtierkunde. G. Fischer. Jena. 1(1): 1-376.
TIBERI, N. 1867. Diagnose du nouveau genre mediterraneen Gyriscus. Journal de Conchyliologie, 15:
303.
TROSCHEL, F.H. 1861. Ueber die systematische Stellung der Gattung Solarium. Archiv fiJ.r Naturgeschichte, 27: 91-99.
TROSCHEL, F.H. 1875. Das Gebiss der Schnecken, zur Begrllndung einer natflrlichen Classification.
Nicolaische Verlagsbuchhandlung. Berlin. 1(4): 133-180.
VAUPEL KLEIN, J.C. von, 1984. A primer of a phylogenetic approach to the taxonomy of the genus
Euchirella {Copepoda, Calanoida). Crustaceana, Supplement 9: 1-194.
VERRILL, A.E. 1881 ["1880"]. Notice of recent additions to the marine Invertebrata, of the northeastern coast of America, with descriptions of new genera and species and critical remarks on others.
Part II. - Mollusca, with notes on Annelida, Echinodermata, etc., collected by the United States
Fish Commission. Proceedings of the United States National Museum, 3: 356-405.
WEBBER, H.H. 1977. Gastropoda: Prosobranchia. Pages 1-97, In: Giese, A.C. & Pearse, J.S. (Eds.),
Reproduction of Marine Int1ertebrates, Vol. IV, Molluscs: Gastropods and Cephalopods. Academic
Press. New York. 369 pp.
WOODRING, W.P. 1928. Miocene mollusks from Bowden, Jamaica. Part II: Gastropods and discussion
of results. Carnegie Institution of Washington Publication, 385: vii, 1-564.
WOODRING, W.P. 1959. Geology and paleontology of Canal Zone and adjoining parts of Panama.
Description of Tertiary mollusks (Gastropods: Vermetidae to Thaididae). United States Geological Survey, Professional Paper, 306-B: iii, 147-239.
RUDIGER BIELER
Smithsonian Marine Station at Link Port, Florida, 33450 U.S.A.
This paper is Contn'bution No.188 of the Smithsonian Marine Station at Link Port, Florida.
Note added in proof: The "unnamed new taxon" cited in text and illustrations has
meanwhile been described as Architectonica (Adelphotectonica) Bieler, 1987, the
subgenus "Heliacus (new taxon)" as Heliacus (Pyrgoheliacus) Bieler, 1987.
)...
APPENDIX
"'1
(")
-·--·
-·
;:t"
TABLE 1. Available data on architectonidd radulae (D =drawing, L =light photomicrograph, S = SEM photomicrograph; "I" marks longest
cusp[s] on rachidian; ..... indicates type species of genus-group taxon).
セ@
(")
0
species studied
radular
formula
no.of teeth
per row
no.of cusps
rachidian
no.of cusps
inner marginal
no.of cusps
outer marginal
;s
reference [note 1)
(")
;::i..
$::>
セ@
Architectonica s.s.
maxima
14-0-14
セ@
28
2
(Philippi, 1849)
;:t"
Boss & Merrill, 1984: 358, pl.63 fig.2 (S),
pl.64 figs.1-3 (S) [as A. perspectiva(note 2)).
セ@
0
OQ
セ@
? perspectiva
(Linne, 1758)
14-0-14
28
• perspectiva
14-0-14
pie ta
1-2
2
Troschel, 1861: 95, pl.5 figs.4-9 (D); 1875:
156, pl.15 figs.4b-f (D); Thiele, 1929: 185
[as Solarium perspectivum (note 3)).
28
2
Boss & Merrill, 1984: 358, pl.56 fig.1 (D)
[not pls.63-64).
14-0-14
28
2
This study (S).
[?]
12-14
[?]
nobilis
7-?-7
14?
?
nobilis
8-0-8
16
n.sp.
7-1-7
15
7-1-7
15
(Philippi, 1849)
nobilis
ROding, 1798
1
2-3
セ@
;s
Bandel, 1984: 62, fig.96 (D); pl.2 fig.4 (S).
2
Boss & Merrill, 1984: 358, pl.56 fig.2,
pl.63 fig.1, pl.65 figs.1-2 (S).
1
2
This study (S).
1
2
2
Boss & Merrill, 1984: 359, pl.66 figs.1-3
(S) [as A. d. laevigata (note 4)).
1+1+1
2
2
Climo, 1975: 281, fig.4B (D).
Architectonica n.subgen.
• reevei
(Hanley, 1862)
"'
(>.)
<.n
TABLE 1 (cont.)
species studied
radular
fonnula
no.of teeth
per row
no.of cusps
rac:hidian
hybrida
(Linne, 1758)
2-1-2
5
1+1+1
3
3
Melone, 1974: 24, pl.2 figs.1-3 (5).
hybrida
2-1-2
5
1+1+1
3
2
Boss & Merrill, 1984: 356, pl50, figs.1-2
(5 by Melone); pl.65 fig.3 (5).
hybrida
2-1-2
5
1+1+1
3-4
3-4
• lutea
(Lamarck, 1822)
2-1-2
5
1+1+1
3
2
Boss & Merrill, 1984: 359, pl57 fig.1 (D).
?
5?
2-3
2-3
2-3
Thiele, 1925: 113 [79), pl.46 (34], fig.18(0).
• krebsii
(Morch, 1875)
2-1-2
5
1+1+1
5
3
Boss & Merrill, 1984: 359, pl57 fig.2 (D).
oxytropis
A. Adams, 1855
2-1-2
5
1+1+1
3
4-5
? radiata
2-1-2
5
2+1+2
4
5
Thiele, 1928: 87, fig.8 (D); 1929: 184,
fig.169 (D) [as Ph. hybrida].
radiata
2-1-2?
5?
?
?
?
Robertson, 1970: 9, figs.9B-O (0).
radiatn
2-1-2
5
1+1+1
5-6
5-6
no.of cusps
inner marginal
no.of cusps
outer marginal
セ@
reference [note 1)
Philippia s.s.
°'
sp.
Merrill, 1970: 31, 204, pl.13 fig.2 (D)
[as Ph. n.sp.]; Melone & Taviani, 1985: 165,
figs.27-28 (S).
Philippia (Psilaxis)
(Roding, 1798)
Climo, 1975: 282, fig.4A (0).
This study (5).
Basisulcata
• lepida
(Bayer, 1942)
セM
セ@
i:::
2-1-2
5
1
7
7
Melone, 1974: 25, pl5 figs.1-2 (5) [as
Acutitectonica mediterranea]; Boss &
Merrill, 1984: 356, pl.53 figs.1-2 (5 by
Melone) [as Acutitectonica hpida].
セ@
セ@
""l
-·セ@
b:J
""l
TABLE 1 (cont)
species studied
lepida
? lepida
radular
fonnula
no.of teeth
per row
no.of cusps
rachidian
2-1-2
5
1
2-1-2
no.of cusps
inner marginal
6-7
no.of cusps
outer marginal
6-7
).
reference [note 1)
"'t
セ@
-·:S-0
--·
Melone & Taviani, 19S5: 161, figs.20-21 (5).
セ@
n:.
0
5
1
SS
SS
Solatisonax
;::t
Boss & Merrill, 1984: 360, pl.5S fig.1 (0)
[as Acutitectonica disca].
-·
セ@
セ@
セ@
n:.
bannocki
2-1-2
5
(Melone & Taviani, 19SO)
borealis
25-30+1+25-30
4-5
4-5
Melone & Taviani, 19S5: 15S, figs.15-16 (5).
2-1-2
5
-12+1+-12
several
several
2-1-2
5
-S+I+-S
5-6
5-6
2-1-2?
5?
-14+1+-14
7
8
cylindricus
2-1-2
5
13-14+1+ 13-14
-5
-5
Boss & Merrill, 19S4: 361, pl.59 fig.1 (0).
trochoides
2-1-2
5
-12+1+-12
6-10?
6-10?
Thiele, 1925: 113 (79), pl.46 (34) fig.16(0).
2-1-2
5
>20+1+>20
9-10
10-13
This study (5).
bisulcatus
2-1-2
5
6-S+l+6-S
6-S
6-8
? implexus
(Mighels, 1845)
2-1-2
5
>20+1+>20
9
9
• implexus [note 6)
2-1-2
(Verrill & Smith, 1SS1)
• injussa
'"a
:S-0
セ@
Iredale, 1931
Boss & Merrill, 1984: 361, pl.60 fig.2 (0)
[as Heliacus (Solatisonax) borealis].
0
OQ
n:.
;::t
セ@
This study (5).
Heliacus s.s.
• cylindricus
(Gmelln, 1791)
Torinia cylindracea]
(Deshayes, 1830)
'f1ariegatus
Troschel, 1S75: 157, figs.7a-c (0) [as
(Gmelin, 1791)
Heliacus (Torinista)
Boss & Merrill, 1984: 362, pl.61 fig.1 (D)
[as Heliacus (Pseudotorinia) bisulcatus].
Habe, 1943: 75, pl.4 fig.5 (D) [as Heliacus
dorsuosus (note 5)).
5
-17+1+-17
N
8-10
8-10
This study (5).
(J.)
'-1
TABLE 1 (cont)
species studied
sterkii
(Pilsbry & Vanatta, 1908)
radular
fonnuJa
no.of teeth
per row
no.of cusps
rachidian
2-1-2
5
15-17+1+15-17
7-9
7-9
This study (5).
2-1-2
5
9+1+9 [note 8)
8-10
8-10
This study (5).
2-1-2
5
18-20+1+18-20
9-10
6
no. of atsps
inner marginal
no.of cusps
outer marginal
セ@
reference [note 1)
00
Heliacus (Grandeliacus)
• stramineus [note 71
(Gmelin, 1791)
suboariegatus
(Orbigny, 1852)
Melone, 1974: 25, pl.4 figs.1-2 (5);
Melone & Taviani, 1985: 167, figs.31-32 (5)
[as H. fallaciosus].
suboariega tus
2-1-2
5
18-22+1+18-22
9-10
9-10
infundibuliformis
(Gmelin, 1791)
2-1-2
5
12-14+1+12-14
-7
-7
This study (5).
• perrieri [note 9)
2-1-2
5
9-10+1+9-10
S7
'S.7
Boss & Merrill, 1984: 361, pl.59 fig.2 (D)
Heliacus (Teretropoma)
Boss & Merrill, 1984: 357, pl.52 figs.1-2 (5
by Melone) [as H. fallaciosus].
[as Heliacus (Heliacus) perrieril.
Heliacus (Gyriscus)
• jeffreysianus
(Tiberi, 1867)
2-1-2
5
10-12+1+10-12
SS
'S.8
Boss & Merrill, 1984: 361, pL60 fig.1 (D).
jeffreysianus
2-1-2
5
12+1+12
4-8
4-8
Melone & Taviani, 1985: 173, fig.41 CD from
Merrill, 1970).
Pseudotorinia [note 10)
architae
2-1-2
(0.G. Costa, 1841)
architae
5
13-15+1+13-15
6
6
Melone, 1974: 24, pl.3 figs.1-2 (5) [as
Heliacus architae].
2-1-2
セ@
セ@
5
-8+1+-8
2-3
2-3
Boss & Merrill, 1984: 362, pl.61 fig.2 (D)
[as Heliacus (Pseudotorinia) architae].
セM
セ@
"'t
b:J
セᄋ@
セ@
"'t
.
TABLE 1 (cont.}
species studied
architae
radular
formula
no.of teeth
per row
no.ofrusps
rachidian
2-1-2
5
15+1+15
no.ofrusps
inner marginal
6
no.ofrusps
outer marginal
6
)..
reference [note 1)
""t
n;:s-
-·--·
-·
nセ@
Melone & Taviani, 1985: 170, figs.36-37 (S)
[as Heliacus (Torinista} architae].
0
;:t
Pseudomalaxis s.s.
n
• zanclaeus zanclaeus
(Philippi, 1844)
2-1-2
zanclaeus zanclaeus
2-1-2
zanclaeus meridionalis
(Hedley, 1903)
2-1-2
5
5
5
S20+1+S20
18-20+1+18-20
20+1+20
4
5-6
6
4
5-6
9
セ@
Boss & Merrill, 1984: 362, pl.62 fig.1 (D)
[as Ps. nobilis].
Melone & Taviani, 1985: 181, figs.56-59 (S).
Kuroda & Habe, 1954: 82, fig.3 (D) [as
Mangonuia solaris].
セ@
セ@
"'a
セ@
セ@
0
O<)
セ@
セ@
zanclaeus meridionlllis
2-1-2
5
-15+1+-15
4
4
This study (S).
• centrifuga
Monterosato, 1890
2-1-2
5
4+1+4
2
2
Boss & Merrill, 1984: 362, pl.62 fig.2 {D)
[as Spirolaxis centrifugal.
centrifuga
2-1-2
5
4-5+1+4-5
2?
2?
Melone & Taviani, 1985: 185, fig.64 (M)
[asSpirolaxis centrifugus1.
Pseudomalaxis (Spirolaxis)
Note 1:
Note 2:
Note 3:
Note 4:
Reproductions of all cited figures published before 1984 can be found in Boss & Merrill (1984), except those of Kuroda & Habe (1954).
Based on USNM 747000, juvenile specimen ('Didi).
Most likely Architectonica perspecti'Oa, but could have included A. trochlearis (Hinds, 1844).
Based on USNM 747441 ('Didi); common but unnamed species, figured as Architectonica sp.III aff. grandiosa Iredale, 1931, by Bieler (1984b: 475,
pl.2 fig.IX).
Note 5: Solarium dorsuosum Hinds, 1844, is considered a nomen dubium (see Bieler, 1985b: 99); HeUacus dorsuosus of authors is H. implexus (Mighels,
1845) in most cases.
Note 6: The type species, Torinista popula Iredale, 1936, is junior subjective synonym of H. implexus (Mighels, 1845} (see Bieler, 1985b: 99).
Note 7: The type species, Grantkliacus mortensenae Iredale, 1957, is junior subjective synonym of H. stramineus (Gmelin, 1791) (see Bieler, 1985b: 100).
Note 8: In this species the central cusp on the rachidian is not longer/stronger than the flanlcing cusps.
Note 9: Heliacus perrieri is here regarded as a subspecies of H. infundibulifonnis (Gmelin, 1791).
Note 10: Pseudotorinia is based on a fossil type species, Solarium obtusum Bronn, 1831.
;:t
Rudiger Bieler
240
TABLE 2. Data matrix used in cladistic analysis ("?"=not known, "X" =not applicable; both
treated as "missing data" in analysis).
Characters:
0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 2 2
1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9
Architectonica
0
0
0
0
0
1
1
0
0
0
0
0
0
newtaxon
Philippia
Psilaxis
Basisulcata
Discoteeton ica
Granosolarium
Solatisonax
Pseudotorinia
Pseudomalaxis
Spirolaxis
Heliacus
OUTGROUP
2 0 1
2 0 1
? 0 ?
1 0 1
? 0 ?
x1 1
x1 1
1 0 1
1 0 1
1 0 1
1 0 1
1 0 1
0 0 0
1 1 0
1 1 0
? ? 0
1 1 0
? ? 0
0 1 1
? ? 1
1 1 0
1 1 0
1 ? 0
1 1 0
1 1 0
0 0 0
2
1
1
1
2
0
0
1
1
0
1
1
1
1
1
1
1
0
0
0
1
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
1
1
1
1
0
0
0
0
0
0
0
0
0
0
0
0
xxxxx
xxxxx
1
1
1
1
1
1
1
1
1
1
1
1
0
1
1
1
1
1
1
0
0
0
0
0
0
x x
2
2
1
2
1
1
1
1
0
1
0
1
0
0
0
0
0
0
0
0
0
0
1
1
0
0
1
1
1
1
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
1
1
1
0
0
0
0
0
0
0
0
0
0
0
1
1
0
0
0
0
1
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
1
0
1
1
1
0
0
1
1
1
1
1
0
0
0
0
0
0
1
0
1
1
1
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
1
1
0
0
0
0
0
1
1
1
1
1
0
0
0
0
0
0
0
0
0
0
0
0
0
1
1
1
0
0
0
0
0
1
1
1
1
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
1
0
0
0
0
0
0
0
0
0
0
0
1
1
0
0
TABLE 3. Cassification of the family Architectonicidae in recent publications ("•" marks taxa
with "rod-like structure" instead of radula).
Garrard (1977):
ARCHITECTONICNAE
Boss &: Merrill (1984):
ARCHITECTONICINAE
Melone &: Taviani (1985):
ARCHITECTONICNAE
Architectonica (Architectonica)
•Architectonica (Discotectonica)
Architectonica (Solatisonax)
Philippia (Philippia)
Philippia (Psilaxis)
Architectonica
Architectonica
Solatisonax
Basisulcata
PHILIPPIINAE
PHILIPPIINAE
Philippia (Philippia)
Philippia (Psilaxis)
•Acutitectonica [note6]
Philippia
PSEUDOMALAXINAE
PSEUDOMALAXINAE
Pseudomalaxis (Pseudomalaxis) [note 1]
Pseudomalaxis (Spirolaxis)
Pseudomalaxis
Spirolaxis
HELIACNAE
Heliacus (Heliacus)
Heliacus (Torinista)
Heliacus (Awarua)
•Heliacus (Claraxis)
[note2]
[note3]
[note4]
[note SJ
HELIACINAE
HELIACINAE
Heliacus [note 7)
Pseudomalaxis
Spirolaxis
A warua [note 4)
Heliacus
Heliacus
Heliacus
Heliacus
Heliacus
(Heliacus)
(Torinista) [note 8)
(Gyriscus)
(Granoheliacus)
(Redivivus)
Note 1: partially included species of Spirolaxis.
Note 2: partially included species of H. (Torinista), H. (Teretropoma) and H. (Grandeliacus).
Note 3: partially included species of H. (Teretropoma) and Pseudotorinia.
Note 4: junior subjective synonym of Pseudotorinia.
Note 5: junior subjective synonym of Granosolarium, but partially included species of Pseudotorinia,
Architectonica and Solatisonax.
Note 6: junior subjective synonym of Discotectonica, but partially included species of Basisulcata.
Note 7: partially included species of Pseudotorinia and Solatisonax.
Note 8: here based on species of Pseudotorinia.
Symposium on Prosobranch Phylogeny, Ninth International Malacological Congress