Available online 6 April 2006
Abstract
The phylogeny of selected members of the phylum Rotifera is examined based on analyses under parsimony direct optimization and
Bayesian inference of phylogeny. Species of the higher metazoan lineages Acanthocephala, Micrognathozoa, Cycliophora, and potential
outgroups are included to test rotiferan monophyly. The data include 74 morphological characters combined with DNA sequence data
from four molecular loci, including the nuclear 18S rRNA, 28S rRNA, histone H3, and the mitochondrial cytochrome c oxidase subunit
I. The combined molecular and total evidence analyses support the inclusion of Acanthocephala as a rotiferan ingroup, but do not sup-
port the inclusion of Micrognathozoa and Cycliophora. Within Rotifera, the monophyletic Monogononta is sister group to a clade con-
sisting of Acanthocephala, Seisonidea, and Bdelloidea?for which we propose the name Hemirotifera. We also formally propose the
inclusion of Acanthocephala within Rotifera, but maintaining the name Rotifera for the new expanded phylum. Within Monogononta,
Gnesiotrocha and Ploima are also supported by the data. The relationships within Ploima remain unstable to parameter variation or to
the method of phylogeny reconstruction and poorly supported, and the analyses showed that monophyly was questionable for the fami-
lies Dicranophoridae, Notommatidae, and Brachionidae, and for the genus Proales. Otherwise, monophyly was generally supported for
the represented ploimid families and genera.
? 2006 Elsevier Inc. All rights reserved.
Keywords: Acanthocephala; Bayesian inference; Cladistics; Rotifera; Hemirotifera; Seison; Syndermata; Micrognathozoa; Cycliophora
1. Introduction
Rotifera is a group of mostly microscopic, aquatic inver-
tebrates with about 1900 described species (Segers, 2002a).
They are generally characterized by the presence of a
corona (or wheel organ) formed of ciliary bands in the
cephalic region. They also have a complex pharyngeal
apparatus, the mastax, composed of hard parts (trophi),
connective musculature, and ligaments (Nogrady et al.,
1993). In a phylogenetic and taxonomic context, the mor-
phology of the trophi is among the most signiWcant charac-
teristics. A total of nine diVerent types of trophi (Fig. 1),
plus some intermediate forms, are recognized.
Three major clades are usually recognized within Rotif-
era: Seisonidea, Bdelloidea, and Monogononta (Wallace
and Colburn, 1989; Wallace and Snell, 1991; Melone et al.,
1998a; S?rensen, 2002). Seisonidea comprises only three
species in the genera Seison and Paraseison (S?rensen et al.,
2005). They are epizoic symbionts on the members of the
leptostracan crustacean genus Nebalia, and are character-
ized by possessing a special fulcrate trophus type (Fig. 1E),
and gamogenetic reproduction (Ricci et al., 1993; Ahlrichs,Molecular Phylogenetics and Evol
A modern approach to rotife
morphological an
Martin V. S?rensen
Department of Organismic and Evolutionary Biology, Museum of
Cambridge, M
Received 30 November 2005; revised 1055-7903/$ - see front matter ? 2006 Elsevier Inc. All rights reserved.
doi:10.1016/j.ympev.2006.04.001
* Corresponding author. Present address: Ancient DNA and Evolution
Group, Niels Bohr Institute and Biological Institute, University of Copen-
hagen, Juliane Maries Vej 30, 2100 Copenhagen, Denmark.
E-mail address: mvsorensen@bi.ku.dk (M.V. S?rensen).ution 40 (2006) 585?608
www.elsevier.com/locate/ympev
ran phylogeny: Combining 
d molecular data
?, Gonzalo Giribet
 Comparative Zoology, Harvard University, 16 Divinity Avenue,
A 02138, USA
6 March 2006; accepted 3 April 20061995a; Segers and Melone, 1998). Bdelloidea comprises ca.
400 species with a rather conserved body plan. The body
consists of a trunk and telescopic retractable head and foot
regions. The head carries a retractable rostrum with adhe-
sive glands used for crawling, and the trophi are of the
ramate type (Fig. 1A). Furthermore, all bdelloids reproduced
586 M.V. S?rensen, G. Giribet / Molecular Phylattempt to evaluate rotiferan relationships without consid-
ering acanthocephalans.ogenetics and Evolution 40 (2006) 585?608solely by parthenogenesis and most species have cryptobi-
otic capabilities (Donner, 1965; Ricci, 1987, 1998a; Mark
Welch and Meselson, 2000; Ricci and Melone, 2000). The
third clade, Monogononta, with ca. 1500 described species,
is by far the most diverse rotifer group and it also displays
the greatest morphological disparity. Monogononts are
mostly free-swimming, even though sessile taxa are repre-
sented as well. The remaining seven trophus types are
found in this clade. Contrary to the exclusively parthenoge-
netic bdelloids, many monogonont rotifers follow a com-
plex cycle composed of a parthenogenetic amictic phase
and a mictic phase that include sexual reproduction and the
presence of haploid dwarf males (Nogrady et al., 1993;
Melone and Ferraguti, 1999).
Also related to rotifers are the members of a parasitic
clade of metazoans, Acanthocephala. While still considered
a phylum in their own, recent molecular data derived from
nuclear ribosomal genes suggest that acanthocephalans are
related to rotifers (Herlyn et al., 2003), perhaps nesting
within them as the sister group to bdelloids (Garc?a-Varela
et al., 2000; Giribet et al., 2004) or sister to Seisonidea
(Zrzav?, 2001a; Herlyn et al., 2003). As such, we cannot
Early microscope pioneers such as van Leeuwenhoek,
Linnaeus, and M?ller originally classiWed rotifers
among the diVuse group ?Vermes? or the protozoan Infu-
soria. Later, Rotifera were assigned to the polyphyletic
Aschelminthes, but more recently an increasing amount
of evidence supports a close relationship between Rotif-
era and Gnathostomulida, and together with Acantho-
cephala and Micrognathozoa they are united in the clade
Gnathifera, supported by the presence of jaws with
identical ultrastructure (Ahlrichs, 1995a,b; Rieger and
Tyler, 1995; Haszprunar, 1996; Giribet et al., 2000; S?ren-
sen et al., 2000; Kristensen and Funch, 2000; Giribet,
2002a; S?rensen, 2003; Funch et al., 2005). The jaws
have been lost secondarily in the highly specialized, endo-
parasitic acanthocephalans, but the presence of a syncy-
tial epidermis with an intraskeletal lamina supports a
close relationship with Rotifera (HaVner, 1950; Koehler,
1966; Storch and Welsch, 1969, 1970). The two
groups have traditionally been considered sister taxa, but
recently the rotifer monophyly with respect to Acantho-
cephala has been questioned. Morphological data
have suggested various combinations of sister-group rela-
Fig. 1. Scanning electron micrographs showing the nine rotiferan trophus types. (A) Ramate trophi of Rotaria neptunia (Bdelloidea), caudal view. (B)
Uncinate trophi of Stephanoceros Wmbriatus (Collothecacea), ventral view. (C) Malleoramate trophi of Filinia longiseta (Flosculariacea), dorsal view. (D)
Malleate trophi of Brachionus plicatilis (Ploima: Brachionidae), ventral view. (E) Fulcrate trophi of Seison nebaliae (Seisonidea), ventral view. (F) Incudate
trophi of Asplanchnopus dahlgreni (Ploima: Asplanchnidae), ventral view. (G) Virgate trophi of Eothinia elongata (Ploima: Notommatidae), right lateral
view. (H) Forcipate trophi of Encentrum astridae (Ploima: Dicranophoridae), ventral view. (I) Cardate trophi of Lindia torulosa (Ploima: Lindiidae).
Abbreviations: fu, fulcrum; ma, manubrium; im, intramalleus; pc, posterior manubrial chamber; ra, ramus; sm, supramanubrium; un, uncus.tionships between Acanthocephala and the three main
rotifer clades, Seisonidea, Bdelloidea, and Monogononta
M.V. S?rensen, G. Giribet / Molecular Phyl
(Lorenzen, 1985; Ahlrichs, 1995a, 1997, 1998; Ferraguti
and Melone, 1999; S?rensen et al., 2000), but all these
have been based on dubious homology statements or con-
tradictory details in sperm anatomy (Melone et al., 1998a;
Ricci, 1998b; Funch et al., 2005). Hence, morphological
data cannot at present provide critical support for Acan-
thocephala as a rotifer ingroup, although a steadily grow-
ing amount of molecular data strongly support its
inclusion (Garey et al., 1996, 1998; Near et al., 1998; Giri-
bet et al., 2000, 2004; Mark Welch, 2000; Miquelis et al.,
2000; Near, 2002).
The discovery of the intriguing Micrognathozoa has
placed a question mark to rotiferan monophyly. Kristensen
and Funch (2000) originally described it as the sister group
to Rotifera and Acanthocephala, whereas De Smet (2002)
interpreted the single micrognathozoan species Limnogna-
thia maerski as an aberrant monogonont rotifer; 18S rRNA
and 28S rRNA sequence data suggested a position closer to
Cycliophora and Gnathostomulida (Giribet et al., 2004).
DiVerent morphological studies have dealt with the
internal phylogeny of Rotifera and most recent work sup-
poses a sister-group relationship between Bdelloidea and
Monogononta, united in the clade Eurotatoria, and place
Seisonidea as their sister group (Wallace and Colburn,
1989; Wallace and Snell, 1991; Melone et al., 1998a; S?ren-
sen, 2002). However, surprisingly few studies have
attempted to resolve the phylogeny within Eurotatoria.
Since 1970, only three studies, Markevich (1989), Melone
et al. (1998a), and S?rensen (2002), have dealt with the rela-
tionships of bdelloid and/or monogonont rotifers, and only
the two latter were based on modern numerical methods.
Melone et al. (1998a) focused solely on bdelloid interrela-
tionships, whereas Markevich (1989) and S?rensen (2002)
included most families of rotifers. Both studies produced
some congruent results, especially concerning the relation-
ships within the diverse and species-rich monogonont clade
Ploima. But they also diVered fundamentally as Markevich
(1989) based his conclusions on the assumption that bdel-
loids present the ancestral character traits, whereas S?ren-
sen (2002) polarized characters based on outgroup
comparison, which placed bdelloids as the sister group to
monogononts. Both studies coincided in the diYculties in
comparing the morphology of the various rotifer taxa and
revealed how the obtained results would be highly depen-
dent on interpretations and character transformations.
In a recent review that summarizes our current knowl-
edge on rotifer systematics, ecology and phylogeny (Wal-
lace, 2002) a call is placed for future studies on rotifer
phylogeny that focus on (1) inclusion of more rotifer taxa in
the analyses, (2) new morphological characters, and (3)
addition of more molecular sequences. In the present study,
we comply with all three demands, as we combine a new
morphological data set with sequences from four molecular
loci representing 53 rotifer species. Our primary goal is to
analyze the relationships between the rotifer main clades
and within Monogononta, but also to test rotifer mono-
phyly with respect to Acanthocephala and Micrognathozoa.ogenetics and Evolution 40 (2006) 585?608 587
2. Materials and methods
2.1. Taxon sampling
The ingroup comprises 48 monogononts, four bdelloids,
four acanthocephalans, Seison nebaliae (Seisonidea), and L.
maerski (Micrognathozoa), spanning most of the diversity
within the phylum Rotifera and potential ingroups. Out-
group taxa include representatives of other gnathiferan and
platyzoan phyla, including Wve Gnathostomulida, four
Platyhelminthes, and the two described species of Cyclio-
phora. Most of the ingroup taxa were collected and
sequenced speciWcally for the present study, while most out-
group sequences were obtained by the authors for two
related studies (Giribet et al., 2004; S?rensen et al., 2006).
Details on collecting localities and GenBank accession
numbers are reported in Table 1.
2.2. DNA sequencing
New DNA sequences were extracted from fresh tissues
using the Qiagen DNeasy? Tissue Kit immediately after
the animals were sorted out. Preferably 10?15 specimens
were used for each extraction, but from samples with a
more restricted number of individuals extractions could
turn out successfully with only 1?3 specimens.
Four molecular loci were chosen for the study. Ribo-
somal sequence data of complete 18S rRNA and a frag-
ment of 28S rRNA were selected to resolve the deeper
nodes in the trees, whereas the nuclear protein-coding gene
histone H3 (H3 hereafter) and the mitochondrial protein-
coding gene cytochrome c oxidase subunit I (COI hereaf-
ter) were included to resolve the more recent evolutionary
events. The complete 18S rRNA loci were ampliWed into
three overlapping fragments using the following primer
pairs: 1F?4R, 3F?18sbi, and 18Sa2.0?9R (Giribet et al.,
1996; Whiting et al., 1997). A fragment of 28S rRNA was
obtained with the primer pairs 28S D1F?28Sb or alterna-
tively 28S D1F?28Sr D4b, which produce fragments of ca.
1200 and 860 bp, respectively (Park and ? Foighil, 2000;
Whiting et al., 1997; Crandall et al., 2000). Histone H3 was
ampliWed with H3aF?H3aR (Colgan et al., 1998), and COI
with LCO1490?HCO2198 (Folmer et al., 1994) or alterna-
tively with COI-7?COI-D (Kojima et al., 1997).
The ampliWed samples were puriWed using QIAquick?
PCR PuriWcation Kit. The puriWed fragments were labeled
using BigDye? Terminator v3.0, and sequenced with an
ABI 3730 genetic analyzer.
Chromatograms obtained from the automated
sequencer were read and ?contig sequences? (assembled
sequences) were assembled using the sequence editing soft-
ware Sequencher? 4.0. For the non-protein-coding genes
(18S rRNA and 28S rRNA), sequences were edited in Mac-
GDE (Linton, 2005) and compared with secondary struc-
ture models and then split into accordant fragments using
internal primers and some visualized secondary structure
features (Giribet and Wheeler, 2001; Giribet, 2002b).
Polyarthra remata Fo588 M.V. S?rensen, G. Giribet / Molecular Phylogenetics and Evolution 40 (2006) 585?608
Table 1
Species represented in the phylogenetic analyses with collecting localities and GenBank accession numbers for each sampled molecular locus
Family Species Collecting locality 18S rRNA 28S rRNA Histone H3 COI
Platyhelminthes Haplopharyngidae Haplopharynx rostratus* ? AJ012511 AF022746 ? AJ405977
Macrostomidae Macrostomum hystricinum* ? AF051329 ? ? ?
Microstomidae Microstomum lineare* ? D85092 AJ270172 ? AJ405979
Notoplanidae Notoplana australis* ? AJ228786 ? ? ?
Cycliophora Symbiidae Symbion americanus Maine and Nova Scotia AY218107 AY218134 AY218154 AY218085
Symbion pandora Sweden AY218106 AY218133 AY218153 AY218084
Gnathostomulida Austrognathiidae Austrognatharia strunki Belize AY218110 AY218137 ? ?
Gnathostomulidae Gnathostomula armata Maine, USA AY218112 AY218139 AY218158 AY218088
Haplognathiidae Haplognathia ruberrima Bermuda DQ079930 DQ079954 AY218156 DQ079968
Mesognathariidae Labidognathia longicollis Belize AY218111 AY218138 AY218087 AY218157
Rastrognathiidae Rastrognathia macrostoma Denmark DQ079935 DQ07959 DQ079988 DQ079970
Micrognathozoa Limnognathiidae Limnognathia maerski Disko Isl., Greenland AY218108 AY218135 AY218155 AY218086
Acanthocephala Echinorhynchidae Echinorhynchus gadi From cod; Disko Isl., 
Greenland
AY218123 AY218146 AY218163 AY218095
Oligacanthorhynchidae Macracanthorhynchus ingens* ? AF001844 ? ? AF416997
Moniliformidae Moniliformis monoliformis* ? Z19562 ? ? AF416998
Pomphorhynchidae Pomphorhynchus laevis From Xounder; Denmark AY218124 ? AY218164 AY218096
Rotifera Seisonidae Seison nebaliae RoscoV, France DQ297761 DQ297762 ? DQ297765
Adinetidae Adineta vaga Cultured by Dr. M. Meselson DQ079913 DQ079940 DQ079974 DQ079961
Philodinidae Philodina acuticornis* U41281 ? ? ?
Rotaria neptunia Denmark AY218122 ? ? AY218094
Rotaria rotatoria Denmark AY218121 ? ? AY218093
Collothecidae Collotheca campanulata Mount Desert Isl., MA, USA DQ297686 DQ297725 ? DQ297766
Conochilidae Conochilus hippocrepis White Mountains, NH, USA DQ297687 DQ297726 DQ297797 DQ297767
Conochilus unicornis White Mountains, NH, USA DQ297688 DQ297727 DQ297798 ?
Filiniidae Filinia longiseta Fort Pierce, FL, USA DQ079914 DQ079941 ? ?
Flosculariidae Ptygura libera Lake Okeechobee, FL, USA DQ297689 DQ297728 ? DQ297768
Sinantherina ariprepes White Mountains, NH, USA DQ297690 DQ297729 ? ?
Testudinellidae Testudinella sp. Mount Desert Isl., MA, USA AY218113 AY218140 AY218159 ?
Gastropodidae Ascomorpha ovalis Mount Desert Isl., MA, USA DQ297691 DQ297730 DQ297799 DQ297769
Asplanchnidae Asplanchnopus dahlgreni Mount Desert Isl., MA, USA DQ079916 DQ079943 DQ079976 DQ079963
Brachionidae Brachionus calyciXorus Cultured by Dr. M. Meselson DQ297692 DQ297731 ? DQ297770
Brachionus plicatilis ? AY218118 AY218144 ? AY218090
Notommatidae Cephalodella forWcula Fort Pierce, FL, USA DQ297693 ? ? ?
Cephalodella gibba Mount Desert Isl., MA, USA AY218114 AY218141 AY218160 AY218089
Dicranophoridae Dicranophorus forcipatus Fort Pierce, FL, USA DQ297694 DQ297732 ? DQ297771
Encentrum astridae Bermuda DQ297695 DQ297733 DQ297800 DQ297772
Encentrum tectipes Bermuda DQ297696 DQ297734 DQ297801 ?
Notommatidae Eothinia elongata Everglades, FL, USA DQ079917 DQ079944 DQ079977 DQ079964
Euchlanidae Euchlanis alata New Hampshire, USA DQ079915 DQ079942 DQ079975 DQ079962
Euchlanis dilatata Mount Desert Isl., MA, USA AY218116 AY218143 ? DQ297773
Brachionidae Keratella quadrata Lake Okeechobee, FL, USA DQ297697 DQ297735 DQ297802 DQ297774
Lecanidae Lecane bulla Fort Pierce, FL, USA DQ297698 DQ297736 DQ297803 DQ297775
Lecane elsa Denmark DQ297699 DQ297737 DQ297804 DQ297776
Lecane leontina Everglades, FL, USA DQ297700 DQ297738 ? DQ297777
Lepadellidae Lepadella patella Fort Pierce, FL, USA DQ297701 DQ297739 DQ297805 DQ297778
Lepadella rhomboides Fort Pierce, FL, USA DQ297702 DQ297740 DQ297806 DQ297779
Lindiidae Lindia tecusa Denmark DQ297703 DQ297741 ? ?
Lindia torulosa Fort Pierce, FL, USA DQ297704 DQ297742 ? ?
Trichotriidae Macrochaetus collinsi Pocono Mountains, PA, USA DQ297705 DQ297743 ? DQ297780
Microcodidae Microcodon clavus Mount Desert Isl., MA, USA DQ297706 DQ297744 DQ297807 DQ297781
Notommatidae Monommata maculata White Mountains, NH, USA DQ297707 DQ297745 DQ297808 ?
Mytilinidae Mytilina mucronata Denmark DQ297708 DQ297746 DQ297809 DQ297782
Mytilina ventralis Fort Pierce, FL, USA DQ297709 DQ297747 DQ297810 DQ297783
Brachionidae Notholca acuminata Winter Harbor, MA, UAS AY218115 AY218142 AY218161 ?
Notommatidae Notommata alantois Lake Okeechobee, FL, USA DQ297710 DQ297748 DQ297811 DQ297784
Notommata cordonella Fort Pierce, FL, USA DQ297711 DQ297749 ? DQ297785
Brachionidae Plationus patulus White Mountains, NH, USA DQ297712 DQ297750 DQ297812 DQ297786
Platyias quadricornis Lake Okeechobee, FL, USA DQ297713 ? ? ?
Synchaetidae Ploesoma hudsoni Mount Desert Isl., MA, USA DQ297714 ? DQ297813 DQ297787
Ploesoma truncatus Pocono Mountains, PA, USA DQ297715 DQ297751 ? DQ297788
rt Pierce, FL, USA DQ297716 DQ297752 DQ297814 DQ297789
M.V. S?rensen, G. Giribet / Molecular Phyl
The protein-coding gene histone H3 showed no length vari-
ation. COI showed some minor length variation and was
analyzed as a single fragment. All new sequences have been
deposited in GenBank under the Accession Nos.
DQ297686?DQ297762 and DQ297765?DQ297820
(Table 1).
2.3. Molecular data
The 18S rRNA fragment was divided into 47 fragments
according to primer regions and secondary structure fea-
tures, as in our previous analyses (e.g., Giribet, 2002b); all
47 fragments were used in the analyses. Sequence length for
the 18S rRNA fragment varied between 1747 nt in Echino-
rhynchus gadi to 1760 nt in Philodina acuticornis, a rather
low sequence length variation. Variation within the out-
groups was larger, between 1717 nt in the gnathostomulid
Gnathostomula armata and 1781 nt in another gnatho-
stomulid, Austrognatharia strunki.
The 28S rRNA fragment of ca. 1 kb also showed 73 nt of
length variation within the ingroup but ranged between 830
and 1119 nt when including outgroups. This gene fragment
was divided into nine regions; the last region measuring
34 bp was excluded from the analysis because it was missing
for more than half of the taxa. Eleven terminals are missing
for the 28S rRNA fragment.
The COI fragment was ampliWed for a total of 53 termi-
nals and due to the presence of length variation (between
651 and 660 nt), it was divided into six regions, aided by the
amino acid translation.
The histone H3 fragment did not ampliWed well for sev-
eral terminals studied and ampliWcation was possible for 40
terminals. The fragment, of 327 nt, showed no length varia-
Table 1 (continued)
All sequences (except those marked with an asterisk) have been generated by 
Family Species Co
Proalidae Proales doliaris M
Proales reinhardti W
Proales similis Be
Scaridiidae Scaridium longicaudum Fo
Trichocercidae Trichocerca elongata M
Trichocerca rattus M
Trichocerca tenuior Fo
Trichotriidae Trichotria tetractis Popatterns have been avoided. During the construction and
coding of the matrix the inclusion of a snumber of charac-ogenetics and Evolution 40 (2006) 585?608 589
ters (generally concerning ultrastructure of sperm and
integument) were considered. These characters could have
resolved some of the basal syndermate splits, but were in
general omitted?either because their character states were
known only for a very restricted number of terminals (three
or less) or because the homology of the characters was con-
sidered uncertain. Further comments to these characters
are given in the following discussion. The list of characters
and their explanations are given in Appendix A and the
morphological matrix is given in Appendix B.
2.5. Data analyses
The morphological data matrix was analyzed with TNT
(Tree analysis using New Technology) (GoloboV et al., 2003)
under the new technology search using a driven search aimed
to stabilize a consensus up to Wve times after Wnding trees of
minimal length multiple times (GoloboV and Farris, 2001;
GoloboV, 2002). This search strategy was compared with a
standard heuristic search strategy consisting of 1000 Wagner
addition trees with subtree pruning and regrafting (SPR) and
tree bisection and reconnection (TBR) branch swapping,
retaining up to 10 trees per replicate.
The molecular data were analyzed using a ?one-step
phylogenetics? approach (Giribet, 2005) in POY v. 3.0.12
(Wheeler et al., 2004) using the direct optimization
method (Wheeler, 1996; Wheeler et al., 2006) with parsi-
mony as the optimality criterion. The data for all the
genes were analyzed independently and in combination.
In addition, the molecular data were analyzed in combi-
nation with the morphological data matrix in POY. Tree
searches were conducted in parallel (using PVM?Paral-
lel Virtual Machine) on a cluster of 30 dual-processor
the authors either for this or for previously published studies.
llecting locality 18S rRNA 28S rRNA Histone H3 COI
ount Desert Isl., MA, USA DQ297717 DQ297753 DQ297815 DQ297790
oods Hole, MA, USA DQ297718 DQ297754 DQ297816 ?
rmuda DQ297719 DQ297755 ? DQ297791
rt Pierce, FL, USA DQ297720 DQ297756 DQ297817 DQ297792
ount Desert Isl., MA, USA DQ297721 DQ297757 DQ297818 DQ297793
ount Desert Isl., MA, USA DQ297722 DQ297758 DQ297819 DQ297794
rt Pierce, FL, USA DQ297723 DQ297759 ? DQ297795
cono Mountains, PA, USA DQ297724 DQ297760 DQ297820 DQ297796tion and was analyzed as a single pre-aligned region.
2.4. Morphological matrix
A morphological matrix containing 74 characters was
compiled from literature sources and from direct observa-
tions of specimens and in particular trophi mounted for
SEM. Preparation of trophi for SEM followed the standard
procedure given by De Smet (1998) and S?rensen (2003).
All terminals are coded in accordance with their actual
morphology, whereas coding based on proposed ground
nodes (between 1 and 2.4 GHz) assembled at Harvard
University (darwin.oeb.harvard.edu). Commands for
load balancing of spawned jobs were in eVect to optimize
parallelization procedures (?parallel ?dpm ?jobspern-
ode 2). Trees were built through a random addition
sequence procedure (20 replicates) followed by a combi-
nation of SPR and TBR branch-swapping, and contin-
uing with tree fusing (TF; GoloboV, 1999, 2002) in order
to further improve tree length. While SPR and TBR
allow branch rearrangement within a given tree, tree
fusing allows exchanging of branches of identical compo-
sition among diVerent trees, as in other simulated evolu-
tionary algorithms (Moilanen, 1999, 2001). Discrepancies
590 M.V. S?rensen, G. Giribet / Molecular Phyl
between heuristic and actual tree length calculations were
addressed by adjusting slop values (?checkslop 10).
While doing tree reWnements using TBR, ?checkslop n
accept all trees that are within n-tenths of a percent of the
current minimum value. For example ?checkslop 10
accepts all trees up to 1% above the current minimum
length while doing TBR.
POY facilitates eYcient sensitivity analysis (Wheeler,
1995; Giribet, 2003). All data sets (individual genes and
combinations) were analyzed under 12 parameter sets, for a
range of indel-to-transversion ratios and transversion-to-
transition ratios (see Table 2). Implied alignments?a topo-
logical-unique ?alignment? or synapomorphy scheme
(Wheeler, 2003; Giribet, 2005)?can be easily generated for
each tree.
After the initial round of 20 replicates with SPR + TBR
+ TF was executed for the 12 parameter sets in the com-
bined analysis of all data, the trees obtained were pooled
and given to POY for a subsequent round of tree fusing, the
so-called sensitivity analysis tree fusing (SATF; Boyer et al.,
2005).
To identify the optimal parameter set, we employed a
character-congruence technique which is a modiWcation of
the incongruence length diVerence (ILD) metric developed
by Mickevich and Farris (1981) (see also Farris et al., 1995),
as proposed by Wheeler (1995) (Table 2). The value is cal-
culated for each parameter set by subtracting the sum of
the scores of all partitions from the score of the combined
analysis of all partitions, and normalizing it for the score of
the combined length. Although the reliability of the ILD
measure employed here has been questioned because it may
Table 2
Tree lengths for the diVerent partitions analyzed (18S, 18S rRNA; 28S
rRNA; H3, histone H3; COI, cytochrome c oxidase subunit I; MOR, mor-
phological data; MOL, four loci combined; TOT, morphology + four loci
combined) and congruence value (ILD) for the combined analysis of mor-
phology + four molecular loci combined at diVerent parameter sets (left
column)
The Wrst numeral used in the parameter set column corresponds to the
ratio between indel/transversion and the following two numbers corre-
spond to the ratio between transversion/transition; e.g., 111 is equal
weights; 121 corresponds to a indel/transversion ratio of 1 and a transver-
sion/transition ratio of 2:1?so indels have a cost of 2, transversions have
a cost of 2, and transitions have a cost of 1. (For a list of the speciWc step
18S 28S H3 COI MOR MOL TOT ILD
111 3736 4232 1277 5148 128 14899 15088 0.037580
121 5600 6413 1890 7990 256 22623 23005 0.037209
141 9146 10483 3058 13461 512 37350 38128 0.038502
181 16226 18474 5344 24341 1024 66664 68159 0.040347
211 4161 4818 1277 5249 256 16039 16425 0.040426
221 6387 7404 1891 8152 512 24688 25414 0.042024
241 10660 12458 3057 13814 1024 41366 42814 0.042066
281 19168 22256 5344 24907 2048 74736 77663 0.050732
411 4767 5589 1277 5305 512 17539 18291 0.045979
421 7550 8868 1890 8269 1024 27652 29111 0.051870
441 12955 15291 3057 14051 2048 47233 50061 0.053115
481 23789 28109 5348 25470 4096 86392 91934 0.055714matrices that this involves, see Giribet et al., 2002, Appendix 4). Optimal
ILD value is indicated in bold.ogenetics and Evolution 40 (2006) 585?608
show a trivial minimum in circumstances in which parti-
tions are given disproportionate weights (Aagesen et al.,
2005), this is not the case here. The modiWed ILD technique
has been interpreted as a meta-optimality criterion for
choosing the parameter set that best explains all partitions
in combination, the one that maximizes overall congruence
and minimizes character conXict among all the data (Giri-
bet, 2003). This parameter set was given special consider-
ation in the analysis of data from each individual gene and
is referred to throughout this paper as the ?optimal param-
eter set.? Additionally, we discuss results from the strict
consensus of all parameter sets explored, which has been
interpreted as a measure of stability to parameter choice, as
applied in statistical sensitivity analyses (Wheeler, 1995;
Giribet, 2003). Nodal support for all topologies was mea-
sured by parsimony jackkniWng (Farris et al., 1996; Farris,
1997).
In order to evaluate the potential eVect of treating each
gap as an independent character (e.g., see Giribet and
Wheeler, 1999; Simmons and Ochoterena, 2000), we run the
combined analysis of all the data under the optimal param-
eter set using a nonlinear (aYne) gap function, where the
gap opening value was higher than that of the gap exten-
sion. This was done by generating the implied alignment for
one of the trees obtained under the optimal parameter set
and using diVerent opening and extension gap costs (2 and
1, respectively). This implied alignment was further used to
perform a Bayesian analysis (Huelsenbeck et al., 2001) in
MrBayes v 3.1.1 (Ronquist and Huelsenbeck, 2005). The
best-Wt model for such implied alignment was calculated
using the Akaike information criterion (AIC) in Modeltest
v. 3.7 (Posada and Crandall, 1998; Posada and Buckley,
2004; Posada, 2005). After selecting the appropriate model
in MrBayes, four chains of 500,000 generations were run
monitoring the potential scale reduction factor (PSRF)
value for convergence. A 95% majority rule consensus tree
was generated with the trees retained after discarding the
burnin.
3. Results
3.1. Morphological data analysis
The driven search strategy performed in TNT gener-
ated a stable consensus containing 41 nodes resulting
from a total of 95 retained trees of 128 steps (Fig. 2). An
identical consensus was obtained after performing 1000
replicates of the traditional heuristic search with SPR and
TBR branch swapping, although this strategy generated
4530 trees and examined 5.8 ? 108 rearrangements as
opposed to the 3.4 ? 107 rearrangements examined with
the Wrst strategy (execution times were 1 and 66 s, respec-
tively, on a Pentium 4 CPU at 2 GHz, 1GB of RAM).
When the same heuristic strategy was implemented
in PAUP* (SwoVord, 2002) in a PowerPC G4 at
1.33 GHz, 1 GB of RAM, the analysis took 751 s (for
9.1 ? 108 rearrangements).
M.V. S?rensen, G. Giribet / Molecular Phylanalysis could not resolve the overall relationships within
Ploima (55% JF). However, the strict consensus tree showsogenetics and Evolution 40 (2006) 585?608 591All trees show monophyly for Syndermata [jackknife
frequency (JF hereafter) <50%] and support a sister-group
relationship between Acanthocephala (99% JF) and Rotif-
era (54% JF). Within Rotifera, Seisonidea is sister group to
Eurotatoria (<50% JF), the latter consisting of Bdelloidea
(91% JF) and Monogononta (<50% JF). Within Monog-
ononta, Collotheca campanulata (viz., the only representa-
tive for Collothecaceae in the analysis) branches oV as the
most basal taxon and forms the sister taxon to a clade con-
sisting of monophyletic Flosculariacea and Ploima. Hence,
monophyly of Gnesiotrocha (consisting of Collothecaceae
and Flosculariaceae) is not supported. The morphological
monophyly for all taxa with malleate trophi, which form a
clade that corresponds to the Transversiramida of Marke-
vich (1989) (see also S?rensen, 2002). Furthermore, mono-
phyly is supported for several clades at the generic and/or
family level, i.e., Lindiidae, Dicranophoridae, Trichocerci-
dae, Lecanidae, Mytilinidae, Euchlanidae, Lepadellidae,
Trichotriidae, and Brachionidae.
3.2. Molecular and total evidence analyses
The combined molecular data were analyzed under
parsimony direct optimization and Bayesian inference.
Fig. 2. Strict consensus of 95 trees at 128 steps obtained after analysis of the morphological data. Numbers on branches indicate jackknife proportions
above 50%. Thicker branches indicate ingroup taxa. Abbreviations for outgroups: PL for Platyhelminthes, CY for Cycliophora, GN for Gnathostomul-
ida, and MI for Micrognathozoa.Subsequently, the combined molecular and morphologi-
cal data were analyzed under direct optimization. For the
592 M.V. S?rensen, G. Giribet / Molecular Phylthe best-Wt model for the implied alignment obtained in
POY. Under this model speciWcation, MrBayes was exe-ogenetics and Evolution 40 (2006) 585?608direct optimization analyses, the congruence measure
identiWed parameter set 121 (indel cost D 2; transversion
cost D 2; transition cost D 1) as the ?optimal? parameter
set (see Table 2). The combined analysis of all molecular
data under this parameter set after SATF resulted in a
single tree of 22,623 weighted steps (Fig. 3). The tree was
almost identical to the resulting trees of the combined
molecular and morphological data analyzed under direct
optimization, hence both results will be discussed together
below.
The AIC implemented in Modeltest chose a general
time-reversible (GTR) model with corrections for the pro-
portion of invariant sites and among-site rate variation as
cuted with two runs of 500,000 generations each until
PSRF approached 1. We discarded 40% of the trees as
burnin, retaining 60,000 trees (30,000 per run). The tree
(Fig. 4) shows syndermate monophyly [posterior proba-
bility (pp hereafter) of 1.00] and divides it into two main
clades consisting of (1) Monogononta (1.00 pp) and (2)
Bdelloidea, Seisonidea, and Acanthocephala (1.00 pp). In
the latter, Seisonidea and Bdelloidea are sister groups
(1.00 pp). Monogononta is divided into Gnesiotrocha
(1.00 pp) and Ploima (1.00 pp). Collotheca campanulata
branches oV inside Flosculariaceae, as sister taxon to Fili-
nia longiseta.
Within Ploima, the two species representing Lepadella
Fig. 3. Single shortest trees at 22,623 weighted steps for the combined analysis of all molecular data analyzed under direct optimization for parameter set
121. Symbols, legends, and abbreviations for outgroups as in Fig. 2.branch oV as the most basal clade. The analysis generally
supports monophyly for all genera, except Trichocerca,
M.V. S?rensen, G. Giribet / Molecular Phyl(tree not shown). Congruence plots ( D Navajo rugs)
for selected relationships involving the major syndermateogenetics and Evolution 40 (2006) 585?608 593Proales, and Cephalodella. Monophyly is supported for
Synchaetidae, but not for Dicranophoridae, Notommati-
dae, Brachionidae, or Trichotriidae (Fig. 4).
The combined analysis of morphological and molecu-
lar data for the optimal parameter set resulted in three
trees of 23,005 weighted steps which consensus is pre-
sented in Fig. 5. Monophyly for Syndermata is supported
under most of the tested parameter sets (Fig. 6), but the
jackknife support for parameter set 121 is less than
50%. In all trees with monophyletic Syndermata, Rotifera
is paraphyletic with respect to Acanthocephala. Micro-
gnathozoa appears within Rotifera only under parameter
set 481 (Fig. 6). In this tree, both Syndermata and
Rotifera are paraphyletic with respect to Gnathostomul-
ida, and Micrognathozoa is sister group to Bdelloidea
clades (Fig. 6) and within Ploima (Fig. 7) are presented
to illustrate the analyses? stability to parameter choice.
Under parameter set 121, Syndermata is divided into
two clades consisting of (1) Monogononta (100% JF) and
(2) Bdelloidea, Seisonidea, and Acanthocephala although
this clade has JF below 50%. Within the latter clade, Bdel-
loidea is sister group to a clade containing Acanthoceph-
ala and Seisonidea. Both Acanthocephala and Bdelloidea
receive jackknife support values near 100%, whereas JF
for Seisonidea + Acanthocephala is below 50% (Figs. 3
and 5).
Monogononta consists of the clades Gnesiotrocha
(85% JF with molecular data; 90% JF with combined
molecular and morphological data) and Ploima (<50%
JF). Collotheca campanulata is nested within Floscularia-
Fig. 4. Fifty percent majority rule consensus of the 60,000 trees retained after discarding the burnin for the Bayesian analysis of phylogeny under a GTR +
G + I model of sequence evolution. Numbers on branches indicate posterior probabilities multiplied by 100. Thicker branches indicate ingroup taxa.
Abbreviations for outgroups as in Fig. 2.ceae, as sister taxon to F. longiseta, hence Flosculariaceae
is paraphyletic. Ploima is monophyletic under all tested
594 M.V. S?rensen, G. Giribet / Molecular Phylbut monophyly is only unambiguously conWrmed for the
latter.ogenetics and Evolution 40 (2006) 585?608parameter sets, and the two species of Lepadella branch
oV as the most basal clade under all parameter sets except
481. Under the optimal parameter set (121), the ploimid
clades branch oV in a ladder-like appearance. However,
only few of these clades obtain JF greater than 50%.
Monophyly is supported for most genera represented by
multiple species. Exceptions are Proales (polyphyletic
under all parameter sets) and Trichocerca (monophyletic
under most parameter sets). Five families, Dicranophori-
dae, Brachionidae, Notommatidae, Trichotriidae, and
Synchaetidae, are represented by more than one genus,
4. Discussion
4.1. Acanthocephalans and their implications for rotifer 
monophyly
Analyses of the combined molecular and morphological
data supported syndermate monophyly under most tested
parameter sets for the direct optimization analyses (Fig. 6),
despite showing jackknife values below 50% (Fig. 5). This is
also the case in other deep nodes such as the one uniting the
non-monogonont forms, or Seisonidea + Acanthocephala.
Fig. 5. Strict consensus of three shortest trees at 23,005 weighted steps for the combined analysis of morphology and molecular data analyzed under direct
optimization for parameter set 121. Symbols, legends, and abbreviations for outgroups as in Fig. 2.The possible disconnection between nodal support and
nodal stability has been explicitly discussed (Giribet, 2003),
M.V. S?rensen, G. Giribet / Molecular Phyl
and stability may be preferred over support by some syste-
matists. Syndermate monophyly is also supported in the
molecular analyses under direct optimization (Fig. 3) as
Fig. 6. Congruence plots (Navajo rugs) for selected relationships involving m
tion ratios) based on the parsimony analyses of the combined morphologica
under the given parameter set. White squares indicate non-monophyly.furthermore admits that aspects of the acanthocephalan
sperm morphology and in particular their mitochondrialogenetics and Evolution 40 (2006) 585?608 595
modiWcations remain unclear, which adds more uncertainty
to the suggested character. Based on the current availability
of data, we preferred not to include the character in our
ajor rotiferan clades (at diVerent gap:change ratios and transversion:transi-
l and molecular data. Black squares indicate monophyly in all trees foundwell as with Bayesian inference of phylogeny (Fig. 4). On
the contrary, monophyly of rotifers, excluding acantho-
cephalans, was only obtained when the morphological data
were analyzed alone (Fig. 2), whereas the molecular data
and combined data set supported acanthocephalans as
being a rotifer ingroup (Figs. 3?6)?i.e., the ?syndermatan
hypothesis.? Rotifer paraphyly with respect to Acantho-
cephala has been debated for years, and nearly all possible
relationships between Acanthocephala and the three rotif-
eran main groups?Bdelloidea, Monogononta, and Seison-
idea?have been proposed (e.g., Wallace, 2002; Herlyn
et al., 2003; Funch et al., 2005).
Scrutiny of morphological characters has not provided
an unambiguous solution to this issue, due to the scarcity of
obviously comparable character traits in the highly special-
ized, endoparasitic acanthocephalans and the morphologi-
cally disparate rotifers. Ahlrichs (1995a,b, 1997) suggested a
sister-group relationship between Acanthocephala and Sei-
sonidea based on the presence of bundles of Wbers in the
mid-zone of the epidermis and dense bodies (probably
mitochondrial derivatives) in the spermatozoa. The latter
character is problematic because studies on sperm ultra-
structure in monogonont rotifers are restricted to species of
the genera Epiphanes, Asplanchna, and Brachionus (Melone
and Ferraguti, 1999), of which only Brachionus plicatilis is
included in the present character matrix. Ahlrichs (1995a)
data matrix for the time being. The similarities between the
Wne structure of integumental details in Seisonidea and
Acanthocephala are certainly a better candidate for a reli-
able morphological character. However, we hesitate to
introduce such a character as it deals with very special
ultrastructural details not necessarily reported in studies
that precede Ahlrichs (1995a) publication. Interpreting a
character trait as ?absent,? solely because it has not been
mentioned in the description, should always be done with
caution. We furthermore have trouble accepting the simi-
larity between the arrangement of the Wbers in Acantho-
cephala and Seisonidea [compare Figs. 6 and 7 in Dunagan
and Miller (1991) with Fig. 2 in Ahlrichs (1997)], hence we
prefer to treat the homology between Wbers in acantho-
cephalans and seisonids as uncertain until more data
become available.
In the study of S?rensen et al. (2000), the absence of a
sperm acrosome was considered synapomorphic for a clade
consisting of Eurotatoria ( D Bdelloidea + Monogononta)
and Acanthocephala. However, as pointed out above,
information about monogonont sperm morphology is
scarce and in our data set restricted to B. plicatilis, hence we
decided not to include the character.
The potential problems with the placement of highly
specialized endoparasites in phylogenetic analyses have
been discussed by Jenner (2004), who pointed out how
extreme morphologies acquired by the parasitic groups
may mislead morphological cladistic analyses. In our case,
596 M.V. S?rensen, G. Giribet / Molecular Phyl2003), Bdelloidea (Garey et al., 1996, 1998; Garc?a-Varela
et al., 2000; Giribet et al., 2000, 2004), or Eurotatoria (Markogenetics and Evolution 40 (2006) 585?608we are explicitly choosing to omit the suggested characters
due to doubtful homology statements or to inapplicability
for most of our terminal species.
The lack of reliable resolution based on morphological
characters has prompted a wealth of molecular evidence to
attempt to resolve rotiferan/acanthocephalan interrelation-
ships. However, these studies have contributed further to
the confusion and proposed acanthocephalans as the sister
group to either Seisonidea (Zrzav?, 2001a; Herlyn et al.,
Welch, 2000; Miquelis et al., 2000). In spite of these incon-
gruities, it is, however, noteworthy that all studies that
involve molecular evidences unambiguously support Acan-
thocephala as a rotifer ingroup (but see Near, 2002). In the
present contribution, we corroborate previous results, test-
ing them by using a variety of analytical conditions and
diVerent phylogenetic approaches and employing the most
comprehensive rotifer taxon sampling so far. Hence, we
Wnd it justiWable to formally include Acanthocephala as a
Fig. 7. Congruence plots (Navajo rugs) for selected relationships involving monogonont taxa based on the parsimony analyses of the combined morpho-
logical and molecular data. Legends and symbols as in Fig. 6.class within the phylum Rotifera. This inclusion makes the
names Rotifera and Syndermata synonyms, Syndermata
M.V. S?rensen, G. Giribet / Molecular Phyl
being the junior synonym. For the sake of stability, it is our
recommendation to use the phylum name Rotifera instead
of its junior synonym Syndermata. It is in this sense that we
use the term Rotifera in this discussion, unless it is other-
wise indicated.
The position of Acanthocephala within Rotifera remains
unsolved. The parsimony analyses supported a sister-group
relationship between Seisonidea and Acanthocephala
under a majority of the tested parameter sets (Figs. 3, 5,
and 6). In this context, it is noteworthy that those parame-
ter sets that reject Seisonidea?Acanthocephala monophyly
mostly coincide with those that do not recognize rotiferan
monophyly and some of those parameters show the highest
amounts of incongruence among data. Thus, the clade
appears to be relatively stable to parameter variation
although jackknife support for a clade of Seisonidea +
Acanthocephala is low, and the analysis under the optimal
parameter has a jackknife support lower than 50% (Fig. 5).
Furthermore, the combined data sets produced a diVerent
result when analyzed with Bayesian inference, and showed
an alternative clade consisting of Bdelloidea + Seisonidea
as the sister group to acanthocephalans (Fig. 4), but with
low posterior probability. In summary, our data appear to
favor a Seisonidea?Acanthocephala relationship, but the
result requires further corroboration.
In spite of the incongruity between the parsimony and
Bayesian analyses, the present results still allow us to evalu-
ate the proposed synapomorphies for both Seisonidea +
Acanthocephala (Ahlrichs, 1995a, 1997) and Eurotatoria +
Acanthocephala (S?rensen et al., 2000), as discussed above.
In all analyses that included molecular data, the reduction
of a sperm acrosome would have been homoplastic, which
supports the questionable nature of the character. The pres-
ence of dense spermatozoon bodies in Acanthocephala and
Seisonidea would, on the other hand, have appeared as a
non-homoplastic character, but the lack of males (and sper-
matozoa) in bdelloid rotifers would render this character as
ambiguously optimized. The presence of Wlamental bundles
in the epidermis of Seisonidea and Acanthocephala is sup-
ported as synapomorphic by the parsimony analyses only
(Figs. 3 and 5), as the Bayesian analysis favors a sister-
group relationship between Seisonidea and Bdelloidea
(Fig. 4). However, ultrastructural data from species of Bdel-
loidea are extremely scarce, and we still prefer to treat this
character with caution until further data from bdelloid spe-
cies have been obtained.
4.2. The rotiferan main clades
The parsimony analysis under the optimal parameter set
and the Bayesian analysis, both supported the division of
Rotifera into two major clades consisting of (1) Monog-
ononta and (2) a clade containing Bdelloidea, Seisonidea,
and Acanthocephala. While L. maerski (Micrognathozoa)
has been suggested to belong to Monogononta (De Smet,
2002), only the analysis under parameter set 481 showed
Micrognathozoa as nested within Rotifera (Fig. 6). Thisogenetics and Evolution 40 (2006) 585?608 597
cladogram also showed Rotifera and Syndermata paraphy-
letic with respect to Gnathostomulida and results in the
highest incongruence of all parameter sets (Table 2).
Instead, based on the congruent results from all other anal-
yses, we Wnd it reasonable to reject L. maerski as part of
Rotifera. This result has been supported in several recent
studies (S?rensen, 2003; Giribet et al., 2004; Funch et al.,
2005).
The clade consisting of Bdelloidea, Seisonidea, and
Acanthocephala was recognized under most parameter
sets (Figs. 5 and 6) and obtains a posterior probability of
1.00 (Fig. 4), hence it can be considered as well supported.
Zrzav? (2001b) suggested the name Lemniscea for this
clade, although the name was originally proposed for a
clade consisting only of Bdelloidea and Acanthocephala
(Garey et al., 1996). We prefer to avoid this name, as the
presence of lemnisci in Bdelloidea and Seisonidea is
doubtful (Ricci, 1998b; see discussion for character 8 in
Appendix A), and introduce the new name Hemirotifera
instead.
All analyses of morphological, molecular, and com-
bined data supported monogonont monophyly, which
agrees with nearly all previous studies (Wallace and Col-
burn, 1989; Nogrady et al., 1993; Melone et al., 1998a;
S?rensen, 2002). The clade is well supported morphologi-
cally, and obtains the highest possible jackknife support
and posterior probability in the combined analyses of all
data (Figs. ?3?5).
4.3. Flosculariaceae and Collothecaceae
The class Monogononta is traditionally divided into the
superorders Pseudotrocha (with the single-order Ploima)
and Gnesiotrocha (Kutikova, 1970; Nogrady et al., 1993;
Segers, 2002a). The latter consists of the two orders Floscu-
lariacea and Collothecacea, and is characterized as a group
of mostly sessile animals with either malleoramate (Fig. 1C)
or uncinate (Fig. 1B) trophi, respectively.
In the present study, Flosculariaceae was represented
with seven terminals and Collothecaceae with a single rep-
resentative, C. campanulata. Whereas all analyses sup-
ported monophyly of Gnesiotrocha, the parsimony
analyses of the combined data only supported the sister-
group relationship between Flosculariacea and Collotheca-
cea under certain parameter sets (Fig. 7). Alternatively, C.
campanulata was positioned as sister group to F. longiseta
(Figs. 3?5). It is, however, likely that this uncertainty is due
to the rather restricted taxon sampling for Collothecacea,
and our data do not seem robust or stable enough for
rejecting the prevalent hypothesis about a sister-group rela-
tionship between the two orders.
Relationships between the included species of Floscular-
iaceae varied with parameter choice and obtained low jack-
knife support under the optimal parameter set (Fig. 5).
This, together with the obvious problems recovering the
position of C. campanulata, shows that the proposed rela-
tionships within the clade are rather unstable, and probably
598 M.V. S?rensen, G. Giribet / Molecular Phyl
not a reXection of the actual relationships. Hence, we do
not Wnd that the illustrated paraphyly of the genus Cono-
chilus (Figs. 3?5) is likely, as a recent study by Segers and
Wallace (2001) shows monophyly of genus.
4.4. Ploima
With ca. 1300 described species, the order Ploima con-
tains more than two-thirds of the total rotifer species. The
order consists of 87 genera, 85 according to Segers (2002a)
plus two described subsequently by De Smet (2003a,b), and
21 families. Generic monophyly is generally considered well
supported within the order, whereas the monophyly is more
questionable for some families, i.e., Proalidae and Notom-
matidae (De Smet, 1996; S?rensen, 2002). Analyses of the
ploimid interrelationships have only been attempted once
using modern numerical methods (S?rensen, 2002), and our
knowledge of this part of the rotifer tree is still extremely
limited.
In all analyses of combined molecular and morphologi-
cal data with the exception of parameter set 481, Lepadelli-
dae, represented with two species of Lepadella, was the
sister group to all remaining Ploima, whose relationships
are extremely parameter-dependant. In all examined trees,
the families Notommatidae and Dicranophoridae come out
as not monophyletic. All parsimony analyses of combined
molecular and morphological data (except under parameter
set 481) supported a clade consisting of Lindia, Monom-
mata, and Dicranophorus (Fig. 5) and under most parame-
ter sets a sister-group relationship between this clade and
Cephalodella was supported (Fig. 7). These aYnities were
also indicated by the Bayesian analysis, even though the
position of Cephalodella could not be fully resolved (Fig. 4).
The remaining representatives of Dicranophoridae, viz.,
two species of Encentrum, show strong aYnities to Proales
doliaris in most analyses, whereas Notommata and Eothinia
(both Notommatidae) often come out together with Asp-
lanchnopus and Synchaetidae (Figs. 4, 5, and 7). The obvi-
ous polyphyly of Notommatidae is not surprising, as the
family has served to include groups that would not Wt in
anywhere else. Hence, its status as non-monophyletic has
already been noted by several authors (e.g., see Koste and
Shiel, 1991; Nogrady, 1995; Segers, 1995; S?rensen, 2002).
Under some of the tested parameter sets, the clade consist-
ing of Cephalodella, Lindia, Monommata, and Dicranopho-
rus branches oV early within Ploima, whereas other
analyses place it as the sister group to a clade consisting of
the remaining Notommatidae and Asplanchnopus. The
restricted taxon sampling of the present analyses does not
allow us to determine whether the notommatid genera are
distantly related and evolved from diVerent positions
within Ploima, or whether they are closely related and con-
stitute part of a clade that also contains taxa such as Asp-
lanchnopus, Lindia, and Synchaetidae. Interesting taxa that
should be added to solve this question would be the species
of the remaining notommatid genera, but probably also
representatives of the genera Tetrasiphon and Itura.ogenetics and Evolution 40 (2006) 585?608
Contrary to the somehow expected notommatid poly-
phyly, it was more surprising to Wnd non-monophyly of
Dicranophoridae. The family appears to be well sup-
ported morphologically (Fig. 2), e.g., by its unique forci-
pate trophus type (Fig. 1H) (see also De Smet, 1997). It is
relatively easy to imagine the transitional series from the
robust trophi in Dicranophorus to the simpler trophi in
Encentrum and closely related genera. However, the tro-
phi in Dicranophorus and other dicranophorid genera, i.e.,
Dorria, also share some basic similarities with the virgate
trophus type (Fig. 1G) and the rami in the cardate trophi
present in Lindia (Fig. 1I). In previous studies, Dicrano-
phorus has been suggested to be one of the most basal
ploimid genera, displaying several plesiomorphic traits,
i.e., the ventral corona (see Remane, 1929?1933). If the
forcipate trophus type likewise is a plesiomorphic trait for
Ploima, it could be considered a precursor for the virgate
trophus type, which would explain the polyphyly of Dicr-
anophoridae.
Brachionidae is the third large ploimid family with
questionable monophyly. The parsimony analysis of the
combined molecular and morphological found support
for its monophyly only under parameter set 411 (Fig. 7),
and in these trees the clade is only supported because of
the inclusion of morphological data. The family is repre-
sented by species of several diVerent genera in the data set,
and in all analyses the genera form two clades consisting
of (1) Brachionidae, Platyias, and Plationus, and (2) Noth-
olca and Keratella. Under some parameter sets, the latter
forms a clade with Proales reinhardti and P. similis,
whereas other parameter sets support a sister-group rela-
tionship with Mytilinidae (Fig. 7). Interestingly, the two
parameter sets support a monophyletic group consisting
of all the brachionid species and Mytilinidae together.
Based on the present results, we Wnd that it is too prema-
ture to reject the possibility of monophyly for Brachioni-
dae, but the position of Mytilinidae and species of Proales
should probably be taken into account in future studies of
the family.
Three species of Trichocerca and one species of Asco-
morpha were included in the data set. These species repre-
sent two well-supported families, Trichocercidae and
Gastropodidae. Previous studies have already demon-
strated some aYnities between the families based on mor-
phological data (S?rensen, 2002). This aYnity was not
conWrmed by the morphological analysis in the present
study (Fig. 2), but all analyses of the combined molecular
and morphological data sets supported a close relation-
ship between these two families. Under most parameter
sets, Ascomorpha and Trichocerca came out as sister
groups (Fig. 7), whereas two parameter sets (121 and 181;
see Fig. 5) and the parsimony and Bayesian analyses of
combined molecular data (Figs. 3 and 4) placed Ascomor-
pha as an ingroup in Trichocerca. We consider that the
suggested paraphyly of Trichocerca is unlikely, because a
sister-group relationship between Trichocercidae and
Gastropodidae appear quite stable to parameter choice.
M.V. S?rensen, G. Giribet / Molecular Phyl
The family Lecanidae is represented by three species of
Lecane, and all analyses conWrmed the monophyly of this
group. Likewise, monophyly was found for Lindiidae,
Mytilinidae, Euchlanidae, and Synchaetidae.
4.5. Future perspectives
Even though the present study represents the most com-
prehensive molecular data set on rotifers assembled so far
and the Wrst combined analysis of molecular and morpho-
logical data, many questions await to be addressed. The data
set corroborates former evidence for the inclusion of Acan-
thocephala within the former Rotifera, in the clade Hemiro-
tifera also containing Bdelloidea and Seisonidea, but the
exact relationship between them still needs further examina-
tion. Future studies should focus on new morphological
characters that could be compared among the free-living
rotifers and the highly modiWed acanthocephalans and sei-
sonideans. Furthermore, more comprehensive molecular
data, e.g., complete sequences of 28S rRNA (see Mallatt and
Winchell, 2002) and the inclusion of more species from these
groups, might produce more conclusive results.
Within Ploima, information about the overall relation-
ships between the families is still limited. However, an
important step forward is to test whether the ploimid fami-
lies represent well-supported monophyletic entities. The
results of the present study do not resolve many questions
concerning the ploimid relationships. Instead, they propose
some possible relationships, candidates for further scrutiny,
and point out some possible non-monophyletic clades at
generic and family levels that should be used with caution
in future studies.
5. Conclusions
Analyses of combined molecular and morphological
data support monophyly of Syndermata and show that
Micrognathozoa is not a member of this clade. Mono-
phyly of Rotifera sensu stricto is not supported by any of
the analyses. Instead, most analyses support the clade
Hemirotifera?consisting of Acanthocephala, Bdelloidea,
and Seisonidea?as sister group to Monogononta. Based
on this evidence and previous congruent results, we for-
mally propose the inclusion of Acanthocephala within
Rotifera.
All analyses support monophyly of Monogononta,
Gnesiotrocha, and Ploima. The sister-group relationship
between the gnesiotrochan orders Flosculariaceae and
Collothecaceae is only supported in a few trees, but this is
most likely an artifact due to insuYcient taxon sampling
for Collothecaceae. The relationships within Ploima are
not resolved, but the analyses indicate that monophyly
should be questioned for the families Dicranophoridae,
Notommatidae, and Brachionidae. The analyses further-
more support a close relationship between Trichocercidae
and Gastropodidae, and conWrmed the monophyly for
most of the included genera and for the families Lepadel-ogenetics and Evolution 40 (2006) 585?608 599
lidae, Lindiidae, Euchlanidae, Lecanidae, Synchaetidae,
and Mytilinidae.
Acknowledgments
Funding to M.V.S. was provided by the Danish Research
Agency (Grant No. 21-04-0331) and by the US National Sci-
ence Foundation. This material is based upon work sup-
ported by the National Science Foundation Assembling the
Tree of Life program under Grant No. 0334932 to G.G. This
is publication #649 from the Smithsonian Marine Station at
Fort Pierce. Associate Editor Stefan Richter and two anony-
mous reviewers provided insightful comments that helped
to improve this publication.
Appendix A. Morphological character description
All characters were coded as non-additive, unless other-
wise indicated.
A.1. General morphology and biology
(1) Epidermis with intraskeletal lamina: 0 D absent,
1 D present.
This character refers to the presence of a proteinaceous
lamina located in the cytoplasm close to the apical mem-
brane of the epidermal cells. It is absent in gnathostomu-
lids, but present in all rotifers and acanthocephalans, as
well as in the dorsal side of L. maerski.
(2) Epidermis structure: 0 D cellular, 1 D syncytial.
In most animals, the integument consists of a cellular
epidermis, but in all rotifer and acanthocephalan species
the epidermis is syncytial. This special condition is reXected
in the name Syndermata Ahlrichs, 1995.
(3) Digestive system: 0 D absent, 1 D present.
The majority of taxa in this analysis possess a fully
developed digestive system, including mouth, pharynx, and
diVerentiated gut. However, in the endoparasitic Acantho-
cephala, the digestive system has been totally reduced.
(4) Animals sessile or mobile: 0 D all stages mobile;
1 D adults sessile (or attached to other individuals in col-
ony); 2 D adults semi-sessile.
The character refers to the locomotory capabilities and
strategies, as these appear to follow some evolutionary
trends. The majority of the taxa in the analysis are mobile
throughout their life history, however, the feeding stages in
Cycliophora are always sessile (Funch and Kristensen,
1995, 1997). This is also true for most rotifers belonging to
Collothecacea and Flosculariacea, as they often reside in
tubes made of pellets or gelatinous material, or live
attached to other individuals in large colonies. Among the
few exceptions are the Xosculariid families Testudinellidae
and Filiniidae, in the analysis represented by Testudinella
patina and F. longiseta, that are both free-swimming. Spe-
cies of Seisonidae stay attached most of their life history,
even though they are capable of detaching and crawl over
for short distances, hence they are coded as semi-sessile.
600 M.V. S?rensen, G. Giribet / Molecular Phyl
The character does not apply to the endoparasitic acantho-
cephalans, hence these are coded as inapplicable.
(5) Animals live in large, free-swimming colony:
0 D absent, 1 D present.
The character refers to the special colonial rotifers, in
this analysis represented by Conochilus hippocrepis and C.
unicornis.
(6) Endoparasites with vertebrate host and intermediate
arthropod host: 0 D absent, 1 D present.
This character refers to a common trait for all acantho-
cephalans, namely their special life cycle that always
involves a vertebrate main host and an arthropod interme-
diate host.
(7) Endoparasite host choice: 0 D Wsh host,
1 D mammalian host.
The acanthocephalan choice of main host may very likely
contain phylogenetically relevant information (Herlyn et al.,
2003), even though former analysis of acanthocephalan rela-
tionships are incongruent (Near et al., 1998; Monks, 2001;
Garcia-Varela et al., 2002; Herlyn et al., 2003).
(8) Lemnisci: 0 D absent, 1 D present.
Lemnisci are paired syncytial, Xuid-Wlled sacks, present
in all acanthocephalans (Hyman, 1951; Ruppert et al.,
2004). Their function is unknown, but they might be
involved in nutrition uptake and Xuid transport. Lorenzen
(1985) suggested a homology between the acanthocephalan
lemnisci and some hypodermic cushions found in certain
bdelloid rotifers. However, Ricci (1998b) demonstrated that
these cushions are present in all rotifers with a well-devel-
oped corona and serve to accommodate the striated ciliary
rootlets. Besides their ectodermal origin, there are no struc-
tural similarities between the lemnisci and the rotifer cush-
ions, hence we follow the view of Ricci (1998b) and code
lemnisci as absent in all rotifers.
(9) Epidermal lacunar system: 0 D absent, 1 D present.
The lacunar system is a special acanthocephalan trait
that consists of a closed channel system in the epidermis
(Hyman, 1951; Ruppert et al., 2004).
(10) With retractile, ciliated rostrum: 0 D absent,
1 D present.
The character refers to the ciliated rostrum present in
most species of Bdelloidea.
(11) Protonephridia: 0 D absent, 1 D present.
Protonephridia are present in all taxa included in the
present study, except three acanthocephalan species.
A.2. Reproductive system
(12) Mode of reproduction: 0 D obligate gamogenetic
reproduction, 1 D obligate parthenogenesis, 2 D  heterogamic
reproduction.
This character refers to the reproductive strategy. Lim-
nognathia maerski is coded as either state 1 or 2, because
males have not yet been recorded. On the other hand, it
cannot be rejected that males might occur during a very
short period. All platyhelminths and gnathostomulids are
hermaphrodites, and their reproduction is gamogenetic.ogenetics and Evolution 40 (2006) 585?608
Cycliophora is coded as gamogenetic as well because dur-
ing a part of their life cycle they reproduce sexually with
separate sexes, whereas neither the inner budding nor the
formation of pandora larvae correspond to the parthenoge-
netic reproduction present in the amictic part of the heter-
ogamic cycle in monogonont rotifers.
(13) Female organs with germovitellarium: 0 D absent,
1 D present.
A syncytial germovitellarium is present in all rotifers
except Seisonidea. The germovitellarium present in some
platyhelminths is not considered homologous with the roti-
fer germovitellarium.
(14) Condition of germovitellarium: 0 D unpaired,
1 D paired.
The rotifer germovitellarium can either be paired, as
present in species of Bdelloidea, or unpaired as present in
monogononts. Taxa without a germovitellarium are coded
an inapplicable for this character.
(15) Life cycle: 0 D dioecious, 1 D hermaphrodite.
The characters distinguish between hermaphroditism
and having separate sexes. Limnognathia maerski is
coded with a question mark, since it is yet uncertain
whether dwarf males are present during a short period of
the life cycle.
(16) Mode of hatching: 0 D oviparous, 1 D viviparous.
Most taxa in the analysis are oviparous, but the bdelloid
species Rotaria neptunia and R. rotatoria carry the embryos
in a brood chamber until they are fully developed (Donner,
1965).
(17) Reproductive organs contained inside collagenous lig-
ament sack(s): 0 D absent, 1 D present.
The reproductive organs in acanthocephalans are con-
tained inside one or more ligament sacks. Such sacks are
not found in any taxa outside Acanthocephala.
(18) Number of ligament sacks: 0 D one, 1 D two; paired
ligament sacks.
Taxa without ligament sacks are coded an inapplicable
for this character.
(19) Female organs with diVerentiated bursa and pre-
bursa: 0 D absent, 1 D present.
The female organs in bursovaginoid gnathostomulids
consist of a sack-shaped organ, which is diVerentiated into
a bursa and a pre-bursa. The bursal system is not found in
taxa outside Bursovaginoidea.
(20) Self-perpetuation by inner budding: 0 D absent,
1 D present.
The character refers to the inner budding found in
feeding stages of Cycliophora. Similar types of budding
are found in certain species of Entoprocta and Ectopr-
octa, but their homology is doubtful (Funch and Kristen-
sen, 1997). Inner budding does not occur in any other taxa
included in this analysis.
A.3. Trophi(21) Jaws composed of translucent rods with an electron-
dense core: 0 D absent, 1 D present.
M.V. S?rensen, G. Giribet / Molecular Phyl
The character refers to the presence of pharyngeal hard
parts composed of longitudinally arranged rods with a con-
spicuous ultrastructure (see Ahlrichs, 1995a; Rieger and
Tyler, 1995; Herlyn and Ehlers, 1997; Kristensen and
Funch, 2000). Such jaws are present only in Gnathostomul-
ida, Micrognathozoa, and Rotifera (but not in Acantho-
cephala).
(22) Fulcrum and rami connected by large hypopharyngeal
muscle working as a piston: 0 D absent, 1 D present.
Most rotifers have a strong hypopharyngeal muscle that
attaches anteriorly to the posterior part of the rami and
posteriorly to the fulcrum or pharyngeal wall. In some taxa,
this muscle is enlarged and pumps like a piston, creating a
vacuum in the pharynx.
(23) Structure of central forceps in jaws: 0 D forceps solely
composed of one layer of parallel rods, 1 D homogenous mass
with tubes inside.
The jaws in Gnathostomulida, Micrognathozoa, and the
non-parasitic Rotifera contain longitudinally arranged rods,
but in the two latter taxa the rods are contained inside a
homogenously appearing material and are thus visible only
in sections. In Gnathostomulida, however, the central forceps
of the jaws, referred to as the articularium (see Riedl and Rie-
ger, 1972), are composed solely of these rods, and the rods
are therefore easily visualized with SEM on isolated hard
parts (S?rensen, 2000; S?rensen and Sterrer, 2002).
(24) General appearance of rami: 0 D rami fan-shaped
composed of small shaft and curved, close-set teeth,
1 D rami short and bulbous, with long projecting alulae
(fulcrate; Fig. 1E), 2 D rami narrow, ribbon-shaped
(ramate; Fig. 1A), 3 D rami extremely thin and delicate
(uncinate; Fig. 1B), 4 D rami elongate, broadly Xattened
with fully opened chambers (malleoramate; Fig. 1C),
5 D rami forceps-shaped with inner margins extended into
1?5 tooth-like processes (incudate; Fig. 1F), 6 D rami
stout, appearing triangular in cross-section (malleate;
Fig. 1D), 7 D rami with apical parts bent characteristically
towards dorsal side (virgate; Fig. 1G), 8 D rami elongate,
Xattened, with large basal chamber Wlling the apical and
basal rami parts (forcipate; Fig. 1H).
The character refers to the general appearance of the
rotifer rami and micrognathozoan main jaws. The rotifer
trophi are traditionally divided into nine types,
namely fulcrate, ramate, uncinate, malleoramate, incudate,
malleate, virgate, cardate, and forcipate. However, since
several intermediate types of trophi occur as well, this char-
acter is restricted to the general appearance of the rami. The
cardate trophus type (Fig. 1I) is mostly deWned by the spe-
cialized manubria (see character 49), whereas the rami
resemble those of the virgate type.
(25) Rami with basal ramus chambers extremely delicate,
forming lateral wing-shaped extensions: 0 D absent,
1 D present.
Rotifer rami are usually thick and solid structures. This
character refers to the extremely delicate rami present in
some taxa, i.e., Microcodon, Ploesoma, Polyarthra, and Syn-
chaeta (latter not represented in analysis).ogenetics and Evolution 40 (2006) 585?608 601
(26) Rami with paired ventral ramus lamellae near midline
of rami: 0 D absent, 1 D present.
The character refers to the paired ventral lamellae, pres-
ent on the rami of Ploesoma, Polyarthra, and Synchaeta.
(27) Anterior rami parts containing basal ramus chambers
isolated, forming antero-dorsal extending projection:
0 D absent, 1 D present.
In species of Cephalodella, the basal ramus chambers tend
to be isolated from the rami, forming a dorsal projection.
(28) Inner margins of rami densely covered with long scle-
ropili: 0 D absent, 1 D present.
Dense set digitiform or needle-shaped structures named
scleropili may occur in various positions in the rotifer tro-
phi. A special feature for species of Flosculariacea is that
the rami inner margins always are densely covered by well-
developed scleropili (Fig. 1C).
(29) Apical parts of rami equipped with strong, curved
band of cristae that Wt with uncus teeth: 0 D absent,
1 D present.
A conspicuous curved band of cristae shaped so that it
interlocks with the unci teeth in the relaxed trophi, is pres-
ent in several taxa.
(30) Ventral sides of rami with anterior processes:
0 D absent, 1 D present.
Anterior processes are isolated sclerites that attach to
the ventral side of the rami through ligaments. The struc-
tures are present in diVerent species of Brachionidae.
(31) Basal apophyses of rami with scleropili: 0 D absent,
1 D present.
The character refers to the presence or absence of scle-
ropili on the basal ramus apophyses, located basally on the
ventral side of the rami.
(32) Each ramus terminates into prominent terminal
tooth: 0 D absent, 1 D present.
In most species of Dicranophoridae, each ramus termi-
nates into a large, well-deWned terminal tooth (Fig. 1H).
Other taxa, i.e., Asplanchnopus (Fig. 1F), may have rami ter-
minating into a tooth-like process, but the character only
codes present for species with a well-deWned terminal tooth.
(33) Rami with slender alulae projecting caudally:
0 D absent, 1 D present.
The character refers to the presence of caudally or lat-
erocaudally projecting alulae (Figs. 1C and E?G). Alulae
are present in various clades within Rotifera (Sanoamuang,
1993; Segers and Wallace, 2001; S?rensen, 2002).
(34) Rami with cardal apophyses interlocking with unci:
0 D absent, 1 D present.
In species of Dicranophoridae and Ituridae (latter not
represented in analysis), the apical parts of the rami have
crests or knobs that interlock with the unci.
(35) Trophi with unci: 0 D absent, 1 D present.
This character refers to the presence of unci (Fig. 1), i.e.,
paired sclerites, often with teeth, located ventroapical to the
rami and proximally articulating with the manubria, forming
the functional units named mallei. Unci are absent in Gna-
thostomulida, but present in L. maerski (Kristensen and
Funch, 2000; De Smet, 2002; S?rensen, 2003), and all jawed
602 M.V. S?rensen, G. Giribet / Molecular Phyl
rotifers included in the analysis, inclusive S. nebaliae (see
Segers and Melone, 1998). In the following characters, the
coding related to the micrognathozoan uncus morphology
follows the interpretation of Kristensen and Funch (2000)
and S?rensen (2003) that suggested a homology between the
rotifer unci and micrognathozoan pseudophalangids.
(36) Apical minor uncus teeth: 0 D absent, 1 D present.
The unci in species of Bdelloidea and Flosculariacea
carry long arrow-like teeth that are diVerentiated into a
median group of broad teeth and groups of apical and/or
sub-basal teeth that are thinner (Sanoamuang, 1993; Segers,
1997a; Melone et al., 1998b; Segers and Wallace, 2001;
Fontaneto et al., 2003). This character refers to the presence
of the apical group of teeth, present in both Bdelloidea and
Flosculariacea.
(37) Sub-basal minor uncus teeth: 0 D absent,
1 D present.
This character refers to the presence of sub-basal minor
uncus teeth, present in Bdelloidea only (Melone et al.,
1998b).
(38) Uncus with scleropilar subuncus: 0 D absent, 1 D present.
This character refers to the presence of an area densely
covered with scleropili on the internal surface of the uncus.
A scleropilar subuncus is present in various rotifer species.
(39) Surface of uncus teeth (non-additive): 0 D no surface
structures, 1 D pore located between uncus head and shaft,
2 D uncus tooth with conspicuous jugal line.
In most species, the uncus teeth have either a distinct
pore or a deep cleft, named jugal line after Markevich
(1989), located on the external surface of each uncus tooth.
However, these structures are absent in S. nebaliae and L.
maerski (Segers and Melone, 1998; S?rensen, 2003). All
bdelloids have a pore in the distal part of each uncus tooth
(Melone et al., 1998b), whereas monogononts have jugal
lines (Markevich, 1989; S?rensen, 2002).
(40) Appearance of proximal part of uncus teeth:
0 D uncus teeth not connected proximally, 1 D uncus teeth
proximally connected by a thin membrane or embedded in
more solid uncus shaft.
The character refers to the diVerence in uncus morphol-
ogy that is observed in species of Bdelloidea and Monog-
ononta. In Bdelloidea, the uncus teeth are not connected
proximally. Instead, each single tooth articulates directly
with the manubria via ligaments (Melone et al., 1998b). In
Gnesiotrocha, the uncus teeth fuse with a thin proximal
uncus membrane, whereas species of Ploima have a well-
developed proximal uncus shaft.
(41) Size of major uncus teeth with jugal lines or pore:
0 D teeth equally sized, 1 D size gradually increasing from
anterior to posterior-most tooth, 2 D one main tooth, others
considerably smaller or reduced to a small plate.
In Bdelloidea and Gnesiotrocha, the major uncus teeth
are equally sized, whereas several species of Ploima have a
large posterior tooth followed by teeth that gradually
decrease in size. Certain ploimid species only have one
prominent uncus tooth next to a plate or a group of consid-
erably smaller teeth.ogenetics and Evolution 40 (2006) 585?608
(42) Trophi with manubria: 0 D absent, 1 D present.
Manubria articulate with the unci and are located lat-
erally in the trophi (Fig. 1). They are present in bdelloids,
monogononts, and in L. maerski but absent in Seisoni-
dea. In the following characters, the coding related to the
micrognathozoan manubrium morphology follows the
interpretation of Kristensen and Funch (2000) and
S?rensen (2003) that suggested a homology between the
rotifer manubria and micrognathozoan accessory
sclerites.
(43) Manubria compartmentalized: 0 D absent, 1 D present.
The character refers to the presence of chambers in the
proximal manubrium heads of monogonont species.
(44) Manubria crescentic with posterior extension:
0 D absent, 1 D present.
The character refers to the morphology of the manubria
in Flosculariacea as coded and described by Segers and
Wallace (2001).
(45) Distal cauda present in manubria: 0 D absent,
1 D present.
The character refers to the presence of a distal rod-
shaped cauda, present in all ploimid species.
(46) Shape of manubrial cauda: 0 D simple rod-shaped,
1 D broadly sickle-shaped.
The character refers to the special sickle-shaped manub-
rial cauda present in species of Brachionus (see also Segers
et al., 1993).
(47) Intramalleus between manubrium and uncus:
0 D absent, 1 D present.
The character refers the presence of a small square- or
droplet-shaped sclerite, the intramalleus, inserted between
the uncus and manubrium (Fig. 1H). The intramalleus
occurs in various species of Dicranophoridae (De Smet,
1997; S?rensen, 2001).
(48) Supramanubrium between manubrium and uncus:
0 D absent, 1 D present.
The character refers the presence of a larger sclerite, the
supramanubrium, between the uncus and manubrium and,
which is characterized as a large projection that points
towards the main axis of the trophi (Fig. 1H). The supr-
amanubrium occurs in various species of Dicranophoridae
(De Smet, 1997; S?rensen, 2001).
(49) Walls of posterior manubrial chamber distally iso-
lated, forming conspicuous appendage on manubrium:
0 D absent, 1 D present.
The character refers to the special manubria that are
present in Lindia (Fig. 1I). In species of Lindia, the walls of
the posterior manubrial chamber form an isolated appen-
dix that only joins the manubrium proximally (e.g., Segers,
2002b).
(50) Right manubrium always strongly reduced; left
manubrium well developed: 0 D absent, 1 D present.
The character refers to the condition in Trichocercidae
where the right manubrium always is strongly reduced (e.g.,
see Sanoamuang and Stout, 1993; Segers, 1997b).(51) Unpaired fulcrum caudal to rami/main jaws:
0 D absent, 1 D present.
M.V. S?rensen, G. Giribet / Molecular Phyl
The character refers to the presence of an unpaired
median structure that extends caudally from the articula-
tion point of the rami/main jaws (Fig. 1). A fulcrum is pres-
ent in Gnathostomulida, Micrognathozoa, Seisonidea, and
Monogononta, but not in Bdelloidea (Melone et al., 1998b;
Segers and Melone, 1998; S?rensen and Sterrer, 2002;
S?rensen, 2002, 2003).
(52) Fulcrum very short and strongly expanded laterally at
distal end: 0 D absent, 1 D present.
The character refers to the peculiar fulcrum present in
species of Lepadellidae. This fulcrum is very short and
strongly expanded laterally at its distal end, giving it a tri-
angular appearance from a dorsal view.
A.4. Ciliation
(53) Ciliation of epidermis: 0 D monociliated,
1 D multiciliated, 2 D non-ciliated.
The character refers to the number of cilia in the epider-
mal cells. Acanthocephalans lack epidermal ciliation,
whereas gnathostomulids have monociliated epidermal
cells. All other taxa in the analysis have multiciliated
epidermis.
(54) Location of cilia from multiciliated cells: 0 D cilia
cover entire animal, 1 D ciliation restricted to head region,
forming a corona consisting of an undivided circumapical
band and a ciliated buccal Weld.
In rotifers, the epidermal ciliation is restricted to the
cephalic region, where cilia are arranged in conspicuous
bands, forming a wheel-organ or corona. The bands may be
modiWed, reduced, or further diVerentiated but generally
the corona is made up by an undivided circumapical band,
apically delimited by the trochus and caudally by the cingu-
lum. The non-ciliated area apical to the trochus is called the
apical Weld, and the often densely ciliated area around the
mouth, inside the circumapical band, is referred to as the
buccal Weld (Remane, 1929?1933; Nogrady et al., 1993).
Limnognathia maerski has cephalic ciliation as well, but the
cilia do not form a corona consisting of an undivided cir-
cumapical band and buccal Weld. Taxa without multicili-
ated epidermal cells are coded as inapplicable for this
character.
(55) Corona type: 0 D Seison type, 1 D Philodina/Adineta
type, 2 D Collotheca type, 3 D Conochilus type,
4 D Hexarthra type, 5 D Euchlanis type, 6 D Notommata/
Dicranophorus type, 7 D Asplanchna type.
The rotifer corona has traditionally been divided into
certain morphological types (Remane, 1929?1933; Ruttner-
Kolisko, 1974). The Seison type is strongly reduced and
only present in the etcoparasitic/commensal Seisonidea.
The corona consists of a small ciliated buccal Weld with two
lateral tufts of cilia. Remane (1929?1933) considered it
close to the Asplanchna type, whereas Ricci et al. (1993) saw
it as closer to the Euchlanis type. Due to these uncertainties
and the special appearance of Seison in general, we prefer
to code the Seison corona in a separate state. The Philodina
type is present in species of the bdelloid order Philodinidaogenetics and Evolution 40 (2006) 585?608 603
(Melone and Ricci, 1995) and consists of a well-developed
ciliary Weld and paired trochi forming ciliary disks. Another
type, the Adineta type, is also present among species of
Bdelloida. This type is much more reduced, but yet closely
related to the Philodina type (Remane, 1929?1933), hence
we chose to code both types in the same state. The Collot-
heca type is present only in the Collothecacea, and is char-
acterized by a completely reduced circumapical band and
buccal Weld ciliation modiWed into a number of stiV tentacle
like bristles. The Conochilus type is formed by a narrow cir-
cumapical band, delimited by well-developed and U-shaped
trochus and cingulum. This type is present in species of the
Xosculariid family Conochilidae. The Hexarthra type, in
older literature referred to as the Pedalia type, is close to
the Conochilus type. It also consists of a narrow circumapi-
cal band, a very well-developed trochus, and a slightly less-
developed cingulum. The apical Weld is large and the rim of
the trochus may form large lobes. The Hexarthra type is
present in all Flosculariacea, exclusive Conochilidae. The
Euchlanis type is dominated by the large buccal Weld, and
parts of it may be modiWed into tufts of cirri called pseudot-
rochi. The circumapical band is a simple band and the rims
are not diVerentiated into trochus and cingulum. The
Euchlanis type is present in various ploimid clades, includ-
ing Brachionidae, Euchlanidae, and Lepadellidae. The Not-
ommata type also consists of a large buccal Weld, and the
circumapical band forms two tufts of large cilia that may be
located on large auricles. The Dicranophorus type is very
close to the Notommata type and is here treated under the
same state. The Asplanchna type consists of a simple cir-
cumapical band without diVerentiated trochus and cingu-
lum. A buccal Weld is absent. This type is usually found in
planktonic ploimids.
(56) Corona with auricles: 0 D absent, 1 D present.
The character refers to the presence of auricles, which
are large, paired rostral extensions densely covered with
long cilia. Auricles are usually present in the Notommata
corona type.
(57) Pseudotroch form tufts with stiV cirri: 0 D absent,
1 D present.
The character refers to the presence of a pseudotroch
with stiV cirri formed by modiWcation of cilia in the buccal
Weld. The pseudotroch is usually present in the Euchlanis
corona type.
(58) Corona with trochal discs on pedicles, separated by
broad upper lip: 0 D absent, 1 D present.
In some species of Bdelloidea, two conspicuous pedicles,
separated by a broad upper lip, extend rostrally from the
head. The pedicles serve as support for the trochal discs
(Melone and Ricci, 1995).
A.5. Lorica
(59) Thickening of integument: 0 D Body illoricate,
1 D Lorica composed of a dorsal and a ventral plate.
2 D Lorica composed of two lateral and a ventral plate,
3 D Lorica fusiform with dorsal ridge(s) on anterior part,
ina may be extremely thin in some species, whereas in other
species it is extraordinary thick, forming heavy plates that
together form a body armor or a so-called lorica. The
developmental grade of the lorica is extremely variable, and
nothing indicates that the rotiferan lorica evolved only
once. However, it is likely that the diVerent lorica shapes
and compositions display some phylogenetic signal. The
present character deWnes Wve diVerent lorica types, mainly
based on the number and location of lorica plates, and the
presence or absence of sulci between the plates.
(60) Appearance of lorica composed of dorsal and ventral
plates: anterior margins on both plates with conspicuous
spines: 0 D absent, 1 D present.
The character refers to the presence or absence of spines
on the anterior margins of the plates in loricas composed of a
dorsal plate and a ventral plate. Taxa coded other than state
1 in character 59 are coded as inapplicable for this character.
(61) Number of spines on dorsal lorica plate: 0 D two
spines, 1 D four spines, 2 D six spines.
The character refers to the number of spines at the ante-
rior margin of the dorsal lorica plate in animals with a
lorica composed of a dorsal plate and a ventral plate. Taxa
coded other than state 1 in character 60 are coded as inap-
plicable for this character.
(62) Dorsal lorica plate with conspicuous ornamentation
forming ridges and notches: 0 D absent, 1 D present.
The character refers to the presence of a conspicuous
ornamentation, forming ridges and lines on the dorsal
lorica plate.
(63) Dorsal and ventral lorica plates connected by broad
and deep sulcus: 0 D absent, 1 D present.
The character refers to the conspicuous deep sulcus
between the dorsal and ventral plates found in species of
Euchlanis.
(64) Ventral lorica plate with distinct transverse fold:
0 D absent, 1 D present.
The character refers to the distinct transverse fold on the
ventral lorica plate, found in species of Lecane.
A.6. Other characters
(65) Caudal foot: 0 D absent, 1 D present.
Most rotifers have a movable foot that extends from the
posterior or in some cases the ventral part of the animal. A
foot may be present in sessile as well as mobile animals.cies of Brachionidae, however, the foot consists of numer-
ous rings and appears distinctively annulated.
(67) Proximal foot pseudosegment diVerentiated into pre-
pedal fold: 0 D absent, 1 D present.
In species of Lecanidae, the most proximal foot,
pseudosegment is modiWed into a pre-pedal fold (see
Segers, 1997c).
(68) Foot pseudosegments covered with heavy, lorica:
0 D absent, 1 D present.
The character refers to the heavily armored foot, present
in species of Trichotriidae.
(69) Foot equipped with toes: 0 D absent, 1 D present.
In most rotifers, the foot terminates into one or usually
two distal toes. In species of Bdelloidea, the number of
toes can be even higher. However, toes are absent in ses-
sile species and in furthermore in a few free-swimming
species as well. In the latter, the foot terminates into a cili-
ated area.
(70) Foot with conspicuous spurs next to toes: 0 D absent,
1 D present.
The character refers to the Xexible paired spurs
near the toes, present in species in species of Trichocerci-
dae.
(71) Toes lengths diVer considerably (ratio > 10:1):
0 D absent, 1 D present.
Rotifer toes are usually equally sized, but in certain spe-
cies of Trichocerca the toes diVer in lengths in a ratio
greater than 10:1.
(72) Toes widest medially: 0 D absent, 1 D present.
Rotifer toes are usually conical or parallel sided, but in
some species of Euchlanis the toes are clearly widest medi-
ally, rather than distally or proximally.
(73) Pedal glands: 0 D absent, 1 D present.
This character refers to the presence of pedal glands,
which are present in all rotifers. Limnognathia maerski
has adhesive glands associated with the ventral cilio-
phores. However, these do not resemble rotiferan pedal
glands, but are more similar to the adhesive glands
found the polychaete Diurodrilus (Kristensen and
Funch, 2000).
(74) Number of pedal glands: 0 D one pair, 1 D multiple
glands.
The character refers to the number of pedal glands that
always are Wxed to two in species of Ploima, whereas it can
be much higher in other rotifers.604 M.V. S?rensen, G. Giribet / Molecular Phylogenetics and Evolution 40 (2006) 585?608
4 D Lorica disk-shaped without sulci,5 D Lorica composed
of one plate with ventral sulcus.
All rotifers possess an integument with an intracellular
Wlamentous layer, called the intraskeletal lamina. This lam-
(66) Appearance of foot: 0 D pseudosegment,
1 D annulated.
The rotifer foot usually consists of one to a few easily
recognizable pseudosegments. In Ploesoma and some spe-
M.V. S?rensen, G. Giribet / Molecular Phylogenetics and Evolution 40 (2006) 585?608 605
Appendix B. Morphological data matrix representing the 74 characters described in Appendix A and coded for the relationships 
of selected rotiferan species plus outgroups
Question marks (?) indicate missing data, dashes (-) indicate inapplicable character states, and A indicates a polymorphism for 1 or 2
1 11 21 31 41 51 61 71
Notoplana australis 001000 - 000 100 - 100 - 00 0- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 10 - - - - - - - - - - 0 - - - - - - - 0 -
Macrostomum hystricinum 001000 - 000 100 - 100 - 00 0 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 10 - - - - - - - - - - 0 - - - - - - - 0 -
Microstomum lineare 001000 - 000 100 - 100 - 00 0 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 10 - - - - - - - - - - 0 - - - - - - - 0 -
Haplopharynx rostratus 001000 - 000 100 - 100 - 00 0 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 10 - - - - - - - - - - 0 - - - - - - - 0 -
Symbion americanus 001100 - 000 100 - 0 - 0 - 01 0 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 10 - - - - - - - - - - 0 - - - - - - - 0 -
Symbion Pandora 001100 - 000 100 - 0 - 0 - 01 0 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 10 - - - - - - - - - - 0 - - - - - - - 0 -
Limnognathia maerski 101000 - 000 1A0 - ?00 - 00 1010 - 0 - 000 000010000 - - 100000000 1010 - - - - 0 - - - - - 0 - - - - - - - 0 -
Haplognathia ruberrima 001000 - 000 100 - 100 - 00 100 - - 0 - - - - 0 - 0 - 0 - - - - - - 0 - - - - - - - - 100 - - - - - - - - - - - 0 - - - - - - - 0 -
Labidognathia longicollis 001000 - 000 100 - 100 - 10 100 - - 0 - - - - 0 - 0 - 0 - - - - - - 0 - - - - - - - - 100 - - - - - - - - - - - 0 - - - - - - - 0 -
Gnathostomula armata 001000 - 000 100 - 100 - 10 100 - - 0 - - - - 0 - 0 - 0 - - - - - - 0 - - - - - - - - 100 - - - - - - - - - - - 0 - - - - - - - 0 -
Rastrognathia macrostoma 001000 - 000 100 - 100 - 10 100 - - 0 - - - - 0 - 0 - 0 - - - - - - 0 - - - - - - - - 100 - - - - - - - - - - - 0 - - - - - - - 0 -
Austrognatharia strunki 001000 - 000 100 - 100 - 10 100 - - 0 - - - - 0 - 0 - 0 - - - - - - 0 - - - - - - - - 100 - - - - - - - - - - - 0 - - - - - - - 0 -
Moniliformis monoliformis 110 - - 11110 000 - 001100 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 2 - - - - - 0 - - - - - 0 - - - - - - - 0 -
Echinorhynchus gadi 110 - - 10110 000 - 001000 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 2 - - - - - 0 - - - - - 0 - - - - - - - 0 -
Pomphorhynchus laevis 110 - - 10110 000 - 001000 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 2 - - - - - 0 - - - - - 0 - - - - - - - 0 -
Macracanthorhynchus ingens 110 - - 11110 100 - 001100 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 2 - - - - - 0 - - - - - 0 - - - - - - - 0 -
Seison nebaliae 111200 - 000 100 - 000 - 00 1111 - 0 - 000 001010000 - - 0 - - - - - - - - 101100000 - - - - - 10000 - - - 11
Adineta vaga 111000 - 000 1111000 - 00 1012 - 0 - 000 0000111010 01000 - 0000 0 - 1110000 - - - - - 100010 0011
Philodina acuticornis 111000 - 000 1111000 - 00 1012 - 0 - 000 0000111010 01000 - 0000 0 - 1110010 - - - - - 100010 0011
Rotaria neptunia 111000 - 000 1111010 - 00 1012 - 0 - 000 0000111010 01000 - 0000 0 - 1110010 - - - - - 100010 0011
Rotaria rotatoria 111000 - 000 1111010 - 00 1012 - 0 - 000 0000111010 01000 - 0000 0 - 1110010 - - - - - 100010 0011
Collotheca campanulata 111100 - 000 1210000 - 00 1013 - 0 - 000 00001000?1 011?0 - 0000 101120000 - - - - - 10000- ?11
Filinia longiseta 111000 - 001 1210000 - 00 1014000100 0010110021 01110 - 0000 101140000 - - - - - 0 - - - 0 - - - 11
Testudinella patina 111000 - 001 1210000 - 00 1014000100 0000110021 01110 - 0000 101140004 - - - - - 10000 - - - 11
Conochilus unicornis 111110 - 001 1210000 - 00 1014000100 0010110021 01110 - 0000 101130000 - - - - - 10000 - - - 11
Conochilus hippocrepis 111110 - 001 1210000 - 00 1014000100 0010110021 01110 - 0000 101130000 - - - - - 10000 - - - 11
Ptygura libera 111100 - 001 1210000 - 00 1014000100 0000110021 01110 - 0000 101140000 - - - - - 10000 - - - 11
Sinantherina ariprepes 111100 - 001 1210000 - 00 1014000100 0000110021 01110 - 0000 101140000 - - - - - 10000 - - - 11
Asplanchnopus dahlgreni 111000 - 001 1210000 - 00 1015000000 0010100021 21?0100000 101170000 - - - - - 100010 0010
Brachionus calyciXorus 111000 - 001 1210000 - 00 1016000011 1000100121 1110110000 1011501011 2000110010 0010
Brachionus plicatilis 111000 - 001 1210000 - 00 1016000011 1000100121 1110110000 1011501011 2000110010 0010
Keratella quadrata 111000 - 001 1210000 - 00 1016000010 1000100121 1110100000 1011501011 21000 - - - - - - - 10
Notholca acuminata 111000 - 001 1210000 - 00 1016000010 1000100121 1110100000 1011501011 30000 - - - - - - - 10
Plationus patulus 111000 - 001 1210000 - 00 1016000011 1000100121 1110100000 1011501011 3100110010 0010
Platyias quadricornis 111000 - 001 1210000-00 1016000010 1000100121 1110100000 1011501011 1100100010 0010
Dicranophorus forcipatus 111000 - 001 1210000 - 00 1018000000 0111100021 2110100000 101160000 - - - - - 100010 0010
Encentrum astridae 111000 - 001 1210000 - 00 1018000000 010110002 - - 110101100 101160000 - - - - - 100010 0010
Encentrum tectipes 111000 - 001 1210000 - 00 1018000000 010110002 - - 110101100 101160000 - - - - - 100010 0010
Euchlanis alata 111000 - 001 1210000 - 00 1016000010 1000100121 1110100000 1011500010 0010100010 0110
Euchlanis dilatata 111000 - 001 1210000 - 00 1016000010 1000100121 1110100000 1011500010 0010100010 0110
Ascomorpha ovalis 111000 - 001 1210000 - 00 1117000000 0010100021 2110100000 101170000 - - - - - 0 - - - - - - - 10
Lecane bulla 111000 - 001 1210000 - 00 1016000000 0000100?21 1110100000 1011500010 0001101010 0010
Lecane elsa 111000 - 001 1210000 - 00 1016000000 0000100?21 1110100000 1011500010 0001101010 0010
Lecane leontina 111000 - 001 1210000 - 00 1016000000 0000100?21 1110100000 1011500010 0001101010 0010
Lepadella patella 111000 - 001 1210000-00 1016000010 1000100?21 1110100000 111150004 - 0 - - - 100010 0010
Lepadella rhomboides 111000 - 001 1210000 - 00 1016000010 1000100?21 1110100000 111150004 - 0 - - - 100010 0010
Lindia tecusa 111000 - 001 1210000 - 00 1018000000 0010100021 2110100010 101161000 - - - - - 100010 0010
Lindia torulosa 111000 - 001 1210000 - 00 1018000000 0010100021 2110100010 101161000 - - - - - 100010 0010
Microcodon clavus 111000 - 001 1210000 - 00 1117100000 0000100021 ?110100000 1011?0000 - - - - - 100010 0010
Mytilina mucronata 111000 - 001 1210000 - 00 1016000010 0000100?21 1110100000 101150002 - - - - 0100010 0010
Mytilina ventralis 111000 - 001 1210000 - 00 1016000010 0000100?21 1110100000 101150002 - - - - 0100010 0010
Cephalodella gibba 111000 - 001 1210000 - 00 1117001000 0000100021 2110100000 101160000 - - - - - 100010 0010
Cephalodella forWcula 111000 - 001 1210000 - 00 1117001000 0000100021 2110100000 101160000 - - - - - 100010 0010
Monommata maculata 111000 - 001 1210000 - 00 1117000000 0010100021 2110100000 101160000 - - - - - 100010 0010
Notommata alantois 111000 - 001 1210000 - 00 1117000000 0010100021 2110100000 101161000 - - - - - 100010 0010
Notommata codonella 111000 - 001 1210000 - 00 1117000000 0010100021 2110100000 101161000 - - - - - 100010 0010
Eothinia elongata 111000 - 001 1210000 - 00 1117000000 0010100021 2110100000 101160000 - - - - - 100010 0010
Proales doliaris 111000 - 001 1210000 - 00 1016000100 0000100121 1110100000 101160000 - - - - - 100010 0010
Proales reinhardti 111000 - 001 1210000 - 00 1016000000 1000100121 1110100000 101160000 - - - - - 100010 0010
Proales similes 111000 - 001 1210000 - 00 1016000100 1010100121 1110100000 101160000 - - - - - 100010 0010
(continued on next page)
De Smet, W.H., 1998. Preparation of rotifer trophi for light and scanning
electron microscopy. Hydrobiologia 387/388, 117?121.
De Smet, W.H., 2002. A new record of Limnognathia maerski Kristen-
sen & Funch, 2000 (Micrognathozoa) from the sub-Antarctic
Crozet Islands, with description of the trophi. J. Zool. Lond. 258,
381?393.
De Smet, W.H., 2003a. Paradicranophorus sinus sp. nov. (Dicranophori-
dae, Monogononta), a new rotifer from Belgium, with remarks on
some other species of the genus Paradicranophorus Wisznievski,
1929 and description of Donneria gen. nov. Belgian J. Morphol. 133,
181?188.
De Smet, W.H., 2003b. Pourriotia carcharodonta, a new genus and species
of monogonont rotifer from sub-Antarctic ?les Kerguelen (Terres
Australes et Antarctiques Fran?aises). Ann. Limnol. Int. J. Limnol. 39,
Garey, J.R., Near, T.J., Nonnemacher, M.R., Nadler, S.A., 1996. Molecular
evidence for Acanthocephala as a subtaxon of Rotifera. J. Mol. Evol.
43, 287?292.
Garey, J.R., Schmidt-Rhaesa, A., Near, T.J., Nadler, S.A., 1998. The evolu-
tionary relationships of rotifers and acanthocephalans. Hydrobiologia
387/388, 83?91.
Giribet, G., 2002a. Current advances in the phylogenetic reconstruction of
metazoan evolution: a new paradigm for the Cambrian explosion?
Mol. Phylogenet. Evol. 24, 409?474.
Giribet, G., 2002b. Relationships among metazoan phyla as inferred
from 18S rRNA sequence data: a methodological approach. In:
DeSalle, R., Giribet, G., Wheeler, W.C. (Eds.), Molecular Systemat-
ics and Evolution: Theory and Practice. Birkh?user Verlag, Basel,
pp. 85?101.606 M.V. S?rensen, G. Giribet / Molecular Phyl
References
Aagesen, L., Petersen, G., Seberg, O., 2005. Sequence length variation,
indel costs, and congruence in sensitivity analysis. Cladistics 21,
15?30.
Ahlrichs, W.H., 1995a. Ultrastruktur und Phylogenie von Seison nebaliae
(Grube 1859) und Seison annulatus (Claus 1876). Cuvillier Verlag, G?t-
tingen. 310 pp..
Ahlrichs, W.H., 1995b. Seison annulatus und Seison nebaliae?Ultrastruk-
tur und Phylogenie. Verhandlungen der Deutschen Zoologischen
Gesellschaft 88, 155.
Ahlrichs, W.H., 1997. Epidermal ultrastructure of Seison nebaliae and Sei-
son annulatus, and a comparison of epidermal structures within Gna-
thifera. Zoomorphology 117, 41?48.
Ahlrichs, W.H., 1998. Spermatogenesis and ultrastructure of the sper-
matozoa of Seison nebaliae (Syndermata). Zoomorphology 118,
255?261.
Boyer, S.L., Karaman, I., Giribet, G., 2005. The genus Cyphophthalmus
(Arachnida, Opiliones, Cyphophthalmi) in Europe: a phylogenetic
approach to Balkan Peninsula biogeography. Mol. Phylogenet. Evol.
36, 554?567.
Colgan, D.J., McLauchlan, A., Wilson, G.D.F., Livingston, S.P., Edge-
combe, G.D., Macaranas, J., Cassis, G., Gray, M.R., 1998. Histone H3
and U2 snRNA DNA sequences and arthropod molecular evolution.
Aust. J. Zool. 46, 419?437.
Crandall, K.A., Harris, D.J., Fetzner Jr., J.W., 2000. The monophy-
letic origin of the freshwater crayWsh estimated from nuclear and
mitochondrial DNA sequences. Proc. R. Soc. B Biol. Sci. 267,
1679?1686.
De Smet, W.H., 1996. The Proalidae (Monogononta). In: Dumont, H.J.F.
(Ed.), Guides to the IdentiWcation of the Microinvertebrates of the
Continental Waters of the World, Vol. 9, Rotifera, Vol. 5: The
Proalidae (Monogononta). SPB Academic Publishing, Amsterdam, pp.
1?102.
De Smet, W.H., 1997. The Dicranophoridae (Monogononta). In: Dumont,
H.J.F. (Ed.), Guides to the IdentiWcation of the Microinvertebrates of
the Continental Waters of the World, Vol. 12, Rotifera, Vol. 5: The
Dicranophoridae (Monogononta). SPB Academic Publishing, Amster-
dam, pp. 1?325.
Appendix B (continued)
1 11 21
Scaridium longicaudum 111000 - 001 1210000 - 00 111700000
Polyarthra remata 111000 - 001 1210000 - 00 111711000
Ploesoma hudsoni 111000 - 001 1210000 - 00 111711000
Ploesoma truncatus 111000 - 001 1210000 - 00 111711000
Trichocerca elongata 111000 - 001 1210000 - 00 111700000
Trichocerca rattus 111000 - 001 1210000 - 00 111700000
Trichocerca tenuior 111000 - 001 1210000 - 00 111700000
Macrochaetus collinsi 111000 - 001 1210000 - 00 101600001
Trichotria tetractis 111000 - 001 1210000 - 00 101600001273?280.ogenetics and Evolution 40 (2006) 585?608
Donner, J., 1965. Ordnung Bdelloidea. Akademie-Verlag, Berlin. 297 pp.
Dunagan, T.T., Miller, D.M., 1991. Acanthocephala. In: Harrison, F.W.
(Ed.), Microscopic Anatomy of Invertebrates, Vol. 4. Wiley-Liss, pp.
299?332.
Farris, J.S., 1997. The future of phylogeny reconstruction. Zool. Scripta 26,
303?311.
Farris, J.S., K?llersj?, M., Kluge, A.G., Bult, C., 1995. Testing signiWcance
of incongruence. Cladistics 10, 315?319.
Farris, J.S., Albert, V.A., K?llersj?, M., Lipscomb, D., Kluge, A.G., 1996.
Parsimony jackkniWng outperforms neighbor-joining. Cladistics 12,
99?124.
Ferraguti, M., Melone, G., 1999. Spermiogenesis in Seison nebaliae (Rotif-
era, Seisonidea): further evidence of a rotifer?acanthocephalan rela-
tionship. Tissue Cell 31, 428?440.
Folmer, O., Black, M., Hoeh, W., Lutz, R., Vrijenhoek, R.C., 1994. DNA
primers for ampliWcation of mitochondrial cytochrome c oxidase sub-
unit I from diverse metazoan invertebrates. Mol. Mar. Biol. Biotechnol.
3, 294?299.
Fontaneto, D., Melone, G., Wallace, R.L., 2003. Morphology of Floscu-
laria ringens (Rotifera, Monogononta) from egg to adult. Invertebr.
Biol. 122, 231?240.
Funch, P., Kristensen, R.M., 1995. Cycliophora is a new phylum with
aYnities to Entoprocta and Ectoprocta. Nature 378, 711?714.
Funch, P., Kristensen, R.M., 1997. Cycliophora. In: Harrison, F.W., Wolla-
cott, R.M. (Eds.), Microscopic Anatomy of Invertebrates, Lophoph-
orates, Entoprocta and Cycliophora, Vol. 13. Wiley-Liss, New York,
pp. 19?39.
Funch, P., S?rensen, M.V., Obst, M., 2005. On the phylogenetic posi-
tion of Rotifera?have we come any further? Hydrobiologia 546,
11?28.
Garc?a-Varela, M., de Leon, G.P.P., de la Torre, P., Cummings, M.P.,
Sarma, S.S.S., Laclette, J.P., 2000. Phylogenetic relationships of Acan-
thocephala based on analysis of 18S ribosomal RNA gene sequences. J.
Mol. Evol. 50, 532?540.
Garcia-Varela, M., Cummings, M.P., Perez-Ponce de Leon, G.,
Gardner, S.L., Laclette, J.P., 2002. Phylogenetic analysis based on
18S ribosomal RNA gene sequences supports the existence of class
Polyacanthocephala (Acanthocephala). Mol. Phylogenet. Evol. 23,
288?292.
31 41 51 61 71
0 0010100021 2110100000 101170000- - - - - 100010 0010
0 0000100001 2110100000 101170000 - - - - - 0 - - - - - - - 10
0 0000100001 2110100000 101170005 - 0- - - 110010 0010
0 0000100001 2110100000 101170005 - 0 - - - 110010 0010
0 0010100021 2110100001 101170003 - 0 - - - 100011 1010
0 0010100021 2110100001 101170003 - 0 - - - 100011 1010
0 0010100021 2110100001 101170003 - 0 - - - 100011 0010
0 ?000100?21 1110100000 1011500010 0100100110 0010
0 1000100?21 1110100000 1011500010 0100100110 0010Giribet, G., 2003. Stability in phylogenetic formulations and its relation-
ship to nodal support. Syst. Biol. 52, 554?564.
M.V. S?rensen, G. Giribet / Molecular Phyl
Giribet, G., 2005. Generating implied alignments under direct optimiza-
tion using POY. Cladistics 21, 396?402.
Giribet, G., Wheeler, W.C., 1999. On gaps. Mol. Phylogenet. Evol. 13,
132?143.
Giribet, G., Wheeler, W.C., 2001. Some unusual small-subunit ribo-
somal RNA sequences of metazoans. Am. Museum Novitates 3337,
1?14.
Giribet, G., Carranza, S., Bagu??, J., Riutort, M., Ribera, C., 1996. First
molecular evidence for the existence of a Tardigrada + Arthropoda
clade. Mol. Biol. Evol. 13, 76?84.
Giribet, G., Distel, D.L., Polz, M., Sterrer, W., Wheeler, W.C., 2000. Triplo-
blastic relationships with emphasis on the acoelomates and the posi-
tion of Gnathostomulida, Cycliophora, Plathelminthes, and
Chaetognatha: a combined approach of 18S rDNA sequences and
morphology. Syst. Biol. 49, 539?562.
Giribet, G., Edgecombe, G.D., Wheeler, W.C., Babbitt, C., 2002. Phylogeny
and systematic position of Opiliones: a combined analysis of chelicer-
ate relationships using morphological and molecular data. Cladistics
18, 5?70.
Giribet, G., S?rensen, M.V., Funch, P., Kristensen, R.M., Sterrer, W., 2004.
Investigations into the phylogenetic position of Micrognathozoa using
four molecular loci. Cladistics 20, 1?13.
GoloboV, P.A., 1999. Analyzing large data sets in reasonable times: solu-
tions for composite optima. Cladistics 15, 415?428.
GoloboV, P.A., 2002. Techniques for analyzing large data sets. In: DeSalle,
R., Giribet, G., Wheeler, W. (Eds.), Techniques in Molecular Systemat-
ics and Evolution. Brikh?user Verlag, Basel, pp. 70?79.
GoloboV, P.A., Farris, J.S., 2001. Methods for quick consensus estimation.
Cladistics 17, 26?34.
GoloboV, P.A., Farris, J.S., Nixon, K., 2003. TNT: tree analysis using new
technology. Version 1.0. Ver. Beta test v. 0.2. Program and documenta-
tion available at http://www.zmuc.dk/public/phylogeny/TNT/.
HaVner, K.V., 1950. Organisation und systematische Stellung der
Acanthocephalen. Zoologischer Anzeiger Suppl. 145, 243?274.
Haszprunar, G., 1996. Platyhelminthes and Platyhelminthomorpha?
paraphyletic taxa. J. Zool. Syst. Evol. Res. 34, 41?48.
Herlyn, H., Ehlers, U., 1997. Ultrastructure and function of the pharynx of
Gnathostomula paradoxa (Gnathostomulida). Zoomorphology 117,
135?145.
Herlyn, H., Piskurek, O., Schmitz, J., Ehlers, U., Zischler, H., 2003. The
syndermatan phylogeny and the evolution of acanthocephalan endo-
parasitism as inferred from 18S rDNA sequences. Mol. Phylogenet.
Evol. 26, 155?164.
Huelsenbeck, J.P., Ronquist, F., Nielsen, R., Bollback, J.P., 2001. Bayesian
inference of phylogeny and its impact on evolutionary biology. Science
294, 2310?2314.
Hyman, L.H., 1951. The Invertebrates: Acanthocephala, Aschelminthes,
and Entoprocta. The Pseudocoelomate Bilateria. McGraw-Hill, New
York.
Jenner, R.A., 2004. Quo Vadis? The Systematist 23, 12?16.
Koehler, J.K., 1966. Some comparative Wne structure relationships of the
Rotifer integument. J. Exp. Zool. 162, 231?243.
Kojima, S., Segawa, R., Hashimoto, J., Ohta, S., 1997. Molecular phy-
logeny of vestimentiferans collected around Japan, revealed by
the nucleotide sequences of mitochondrial DNA. Mar. Biol. 127,
507?513.
Koste, W., Shiel, R.J., 1991. Rotifera from Australian inland waters. VII.
Notommatidae (Rotifera: Monogononta). Trans. R. Soc. South Aust.
115, 111?159.
Kristensen, R.M., Funch, P., 2000. Micrognathozoa: a new class with com-
plicated jaws like those of Rotifera and Gnathostomulida. J. Morphol.
246, 1?49.
Kutikova, L.A., 1970. The Rotifer Fauna of USSR. Nauka Publishers,
Leningrad. 744 pp.
Linton, E.W., 2005. MacGDE: genetic data environment for MacOS X.
Ver. 2.0. Software available at www.msu.edu/ ~ lintone/macgde/.Lorenzen, S., 1985. Phylogenetic aspects of pseudocoelomate evolution. In:
Conway-Morris, S., George, J.D., Gibson, R., Platt, H.M. (Eds.), Theogenetics and Evolution 40 (2006) 585?608 607
Origins and Relationships of Lower Invertebrates: Systematic Associa-
tion. Clarendon Press, Oxford, pp. 210?223.
Mallatt, J., Winchell, C.J., 2002. Testing the new animal phylogeny: Wrst
use of combined large-subunit and small-subunit rRNA gene
sequences to classify the protostomes. Mol. Biol. Evol. 19, 289?301.
Mark Welch, D.B., 2000. Evidence from a protein-coding gene that
acanthocephalans are rotifers. Invertebr. Biol. 119, 17?26.
Mark Welch, D.B., Meselson, M., 2000. Evidence for the evolution of bdel-
loid rotifers without sexual reproduction or genetic exchange. Science
288, 1211?1215.
Markevich, G.I., 1989. Morphology and principal organization of sclerite sys-
tem of the mastax in rotifers. In: Shilova, A.I. (Ed.), Biological and Func-
tional Morphology of Freshwater Animals, Proceedings of the Institute of
the Biology of Inland Waters, Leningrad 56, pp. 27?82 (in Russian).
Melone, G., Ferraguti, M., 1999. Rotifera. In: Jamieson, B.G.M. (Ed.),
Reproductive Biology of Invertebrates, Part A, Progress in Male Gam-
ete Biology, Vol. IX. Oxford and IBH Publishing Co., New Delhi, pp.
157?169.
Melone, G., Ricci, C., 1995. Rotatory apparatus in bdelloids. Hydrobiolo-
gia 313/314, 91?98.
Melone, G., Ricci, C., Segers, H., Wallace, R.L., 1998a. Phylogenetic rela-
tionships of phylum Rotifera with emphasis on the families of Bdelloi-
dea. Hydrobiologia 387/388, 101?107.
Melone, G., Ricci, C., Segers, H., 1998b. The trophi of Bdelloidea (Rotif-
era): a comparative study across the class. Can. J. Zool. 76, 1755?1765.
Mickevich, M.F., Farris, J.S., 1981. The implications of congruence in
Menidia. Syst. Zool. 27, 143?158.
Miquelis, A., Martin, J.F., Carson, E.W., Brun, G., Gilles, A., 2000. Perfor-
mance of 18S rDNA helix E23 for phylogenetic relationships within
and between the Rotifera?Acanthocephala clades. Comptes Rendus de
l?Academie Des Sciences. Serie III, Sciences de la Vie 323, 925?941.
Moilanen, A., 1999. Searching for most parsimonious trees with simulated
evolutionary optimization. Cladistics 15, 39?50.
Moilanen, A., 2001. Simulated evolutionary optimization and local search:
introduction and application to tree search. Cladistics 17, 12?25.
Monks, S., 2001. Phylogeny of the Acanthocephala based on morphologi-
cal characters. Syst. Parasitol. 48, 81?116.
Near, T.J., 2002. Acanthocephalan phylogeny and the evolution of parasit-
ism. Integr. Comp. Biol. 42, 668?677.
Near, T.J., Garey, J.R., Nadler, S.A., 1998. Phylogenetic relationships of
the Acanthocephala inferred from 18S ribosomal DNA sequences.
Mol. Phylogenet. Evol. 10, 287?298.
Nogrady, T., 1995. Notommatidae. In: Dumont, H.J.F. (Ed.), Guides to the
IdentiWcation of the Microinvertebrates of the Continental Waters of
the World, Vol. 8, Rotifera, Vol. 3: Notommatidae. SPB Academic
Publishing, Amsterdam, pp. 1?229.
Nogrady, T., Wallace, R.L., Snell, T.W., 1993. In: Nogrady, T. (Ed.),
Guides to the IdentiWcation of the Microinvertebrates of the Continen-
tal Waters of the World, Vol. 4, Rotifera, Vol. 1: Biology, Ecology and
Systematics. SPB Academic Publishing, Amsterdam, pp. 1?142.
Park, J.K., ? Foighil, D., 2000. Sphaeriid and corbiculid clams represent
separate heterodont bivalve radiations into freshwater environments.
Mol. Phylogenet. Evol. 14, 75?88.
Posada, D., 2005. Modeltest 3.7. Program and documentation available at
darwin.uvigo.es.
Posada, D., Buckley, T., 2004. Model selection and model averaging
in phylogenetics: advantages of Akaike information criterion and
Bayesian approaches over likelihood ratio tests. Syst. Biol. 53,
793?808.
Posada, D., Crandall, K.A., 1998. MODELTEST: testing the model of
DNA substitution. Bioinformatics 14, 817?818.
Remane, A., 1929?1933. Rotatoria, In: Bronn, H.G. (Ed.), Klassen und
Ordnungen des Tier-Reichs, Vol. 4, Vermes. Akademische Verlagsge-
sellschaft m. b. H., Leipzig, pp. 1?576.
Ricci, C., 1987. Ecology of bdelloids: how to be successful. Hydrobiologia
147, 117?127.Ricci, C., 1998a. Anhydrobiotic capabilities of bdelloid rotifers. Hydrobio-
logia 387/388, 321?326.
608 M.V. S?rensen, G. Giribet / Molecular Phylogenetics and Evolution 40 (2006) 585?608
Ricci, C., 1998b. Are lemnisci and proboscis present in the Bdelloidea?
Hydrobiologia 387/388, 93?96.
Ricci, C., Melone, G., 2000. Key to the identiWcation of the genera of bdel-
loid rotifers. Hydrobiologia 418, 73?80.
Ricci, C., Melone, G., Sotgia, C., 1993. Old and new data on Seisonidea
(Rotifera). Hydrobiologia 255/256, 495?511.
Riedl, R., Rieger, R., 1972. New characters observed on isolated jaws and
basal plates of the family Gnathostomulidae (Gnathostomulida). Z.
Encentrum (Rotifera: Monogononta: Ploima). Proc. Biol. Soc. Wash-
ington 114, 725?736.
S?rensen, M.V., 2002. On the evolution and morphology of the rotiferan
trophi, with a cladistic analysis of Rotifera. J. Zool. Syst. Evol. Res. 40,
129?154.
S?rensen, M.V., 2003. Further structures in the jaw apparatus of Limno-
gnathia maerski (Micrognathozoa), with notes on the phylogeny of the
Gnathifera. J. Morphol. 255, 131?145.Morph. Tiere 72, 131?172.
Rieger, R.M., Tyler, S., 1995. Sister-group relationship of Gnathostomul-
ida and Rotifera-Acanthocephala. Invertebr. Biol. 114, 186?188.
Ronquist, F., Huelsenbeck, J.P., 2005. MrBayes v3.1.1: Bayesian analysis
of phylogeny. Program and documentation available at: http://mrba-
yes.csit.fsu.edu/index.php.
Ruppert, E.E., Fox, R.S., Barnes, R.D., 2004. Invertebrate Zoology.
Brooks/Cole?Thomson Learning, Belmont, CA.
Ruttner-Kolisko, A., 1974. Plankton rotifers: biology and taxonomy. Die
Binnengew?sser Suppl. 16, 1?146.
Sanoamuang, L., 1993. Comparative studies on scanning electron micros-
copy of the trophi of the genus Filinia Bory De St. Vincent (Rotifera).
Hydrobiologia 264, 115?128.
Sanoamuang, L., Stout, V.M., 1993. New records of rotifers from the
South Island lakes, New Zealand. Hydrobiologia 255/256, 481?490.
Segers, H., 1995. A reappraisal of the Scaridiidae (Rotifera, Monog-
ononta). Zool. Scripta 24, 91?100.
Segers, H., 1997a. Contribution to a revision of Floscularia Cuvier, 1978
(Rotifera: Monogononta): notes on some Neotropical taxa. Hydrobio-
logia 354, 165?175.
Segers, H., 1997b. Some Rotifera from the collection of the Academy of
Natural Science of Philadelphia, including new species and new
records. Proc. Acad. Natl. Sci. Philadelphia 148, 147?156.
Segers, H., 1997c. The Lecanidae (Monogononta). In: Dumont, H.J.F.
(Ed.), Guides to the IdentiWcation of the Microinvertebrates of the
Continental Waters of the World, Vol. 6, Rotifera, Vol. 2: The
Lecanidae (Monogononta). SPB Academic Publishing, Amsterdam,
pp. 1?226.
Segers, H., 2002a. The nomenclature of the Rotifera: annotated checklist
of valid family- and genus-group names. J. Nat. History 36, 631?640.
Segers, H., 2002b. Family Lindiidae. In: Nogrady, T., Segers, H. (Eds.),
Guides to the IdentiWcation of the Microinvertebrates of the Continen-
tal Waters of the World, Vol. 18, Rotifera, Vol. 6: Asplanchnidae, Gas-
tropodidae, Lindiidae, Microcodidae, Synchaetidae, Trochosphaeridae
and Filinia. Backhuys Publishers, Leiden, pp. 55?82.
Segers, H., Melone, G., 1998. A comparative study of trophi morphology
in Seisonidea (Rotifera). J. Zool. Lond. 244, 201?207.
Segers, H., Wallace, R.L., 2001. Phylogeny and classiWcation of the Conochi-
lidae (Rotifera, Monogononta, Flosculariacea). Zool. Scripta 30, 37?48.
Segers, H., Murugan, G., Dumont, H.J., 1993. On the taxonomy of the
Brachionidae: description of Plationus n. gen. (Rotifera, Monog-
ononta). Hydrobiologia 268, 1?8.
Simmons, M.P., Ochoterena, H., 2000. Gaps as characters in sequence-
based phylogenetic analyses. Syst. Biol. 49, 369?381.
S?rensen, M.V., 2000. An SEM study of the jaws of Haplognathia rosea
and Rastrognathia macrostoma (Gnathostomulida), with a preliminary
comparison with rotiferan trophi. Acta Zool. 81, 9?16.
S?rensen, M.V., 2001. The rotifer fauna of Bermuda, including notes on
the associated meiofauna and the description of a new species ofS?rensen, M.V., Sterrer, W., 2002. New characters in the gnathostomulid
mouth parts revealed by scanning electron microscopy. J. Morphol.
253, 310?334.
S?rensen, M.V., Funch, P., Willerslev, E., Hansen, A.J., Olesen, J., 2000. On
the phylogeny of the Metazoa in the light of Cycliophora and Micro-
gnathozoa. Zoologischer Anzeiger 239, 297?318.
S?rensen, M.V., Funch, P., Segers, H., 2005. On a new Seison Grube, 1861
from coastal waters of Kenya, with a reappraisal of the classiWcation of
Seisonida (Rotifera). Zool. Stud. 44, 34?43.
S?rensen, M.V., Sterrer, W., Giribet, G., 2006. Gnathostomulid phylogeny
inferred from a combined approach of four molecular loci and mor-
phology. Cladistics 22, 32?58.
Storch, V., Welsch, U., 1969. ?ber den Aufbau des Rotatorienintegumen-
tes. Zeitschrift f?r Zellforschung 95, 405?414.
Storch, V., Welsch, U., 1970. ?ber den Aufbau resorbierender Epithelien
darmloser Endoparasiten. Zoologischer Anzeiger Suppl. 33, 617?621.
SwoVord, D.L., 2002. PAUP* 4.0: Phylogenetic Analysis Using Parsimony
(* and Other Methods). Ver. 4. Sinauer Associates.
Wallace, R.L., 2002. Rotifers: exquisite metazoans. Integr. Comb. Biol. 42,
660?667.
Wallace, R.L., Colburn, R.A., 1989. Phylogenetic relationships within phy-
lum Rotifera?orders and genus Notholca. Hydrobiologia 186, 311?318.
Wallace, R.L., Snell, T.W., 1991. Rotifera. In: Thorpe, J., Covich, A. (Eds.),
Ecology and ClassiWcation of North American Freshwater Inverte-
brates. Academic Press, New York, pp. 187?248.
Wheeler, W.C., 1995. Sequence alignment, parameter sensitivity, and the
phylogenetic analysis of molecular data. Syst. Biol. 44, 321?331.
Wheeler, W.C., 1996. Optimization alignment: the end of multiple
sequence alignment in phylogenetics? Cladistics 12, 1?9.
Wheeler, W.C., 2003. Implied alignment: a synapomorphy-based multiple-
sequence alignment method and its use in cladogram search. Cladistics
19, 261?268.
Wheeler, W.C., Gladstein, D., De Laet, J., 2004. POY version 3.0. Ver. Pro-
gram and documentation available at ftp.amnh.org/pub/molecular.
American Museum of Natural History.
Wheeler, W.C., Aagesen, L., Arango, C.P., Faivovich, J., Grant, T., D?Haese,
C., Janies, D., Smith, W.L., Varon, A., Giribet, G., 2006. Dynamic
Homology and Phylogenetic Systematics: A UniWed Approach using
POY. American Museum of Natural History, New York.
Whiting, M.F., Carpenter, J.M., Wheeler, Q.D., Wheeler, W.C., 1997. The
Strepsiptera problem: phylogeny of the holometabolous insect orders
inferred from 18S and 28S ribosomal DNA sequences and morphol-
ogy. Syst. Biol. 46, 1?68.
Zrzav?, J., 2001a. The interrelationships of metazoan parasites: a review of
phylum- and higher-level hypothesis from recent morphological and
molecular phylogenetic analyses. Folia Parasitol. 48, 81?103.
Zrzav?, J., 2001b. Myzostomida are not annelids: molecular and morpho-
logical support for a clade of animals with anterior sperm Xagella. Cla-
distics 17, 170?198.