ArticlePDF Available

The Earliest Endosymbiotic Mineralized Tubeworms from the Silurian of Podolia, Ukraine


Abstract and Figures

The earliest endosymbiotic tubeworms have been discovered within skeletons of the tabulate coral Heliolites sp. from the Silurian (Ludlow) of Podolia, Ukraine. The new tubeworm species has a maximum diameter about I mm, a slightly conical tube, a smooth lumen in the tube and a lamellar wall structure. The tube wall is 0.05-0.10 mm thick. The new endosymbiotic tubeworm Coralloconchus bragensis n. gen. and sp. shares zoological affinities with the tentaculitids (incertae sedis) and is assigned to the Family Cornulitidae (Tentaculita, Comulitida).
Content may be subject to copyright.
J. Paleont., 82(2), 2008, pp. 409–414
Copyright 2008, The Paleontological Society
Institute of Geology, University of Tartu, Vanemuise 46, 51014 Tartu, Estonia,;
Institute of Geology, Tallinn University of
Technology, Ehitajate str. 5, 19086 Tallinn, Estonia,
—The earliest endosymbiotic tubeworms have been discovered within skeletons of the tabulate coral Heliolites sp. from the Silurian
(Ludlow) of Podolia, Ukraine. The new tubeworm species has a maximum diameter about 1 mm, a slightly conical tube, a smooth lumen in
the tube and a lamellar wall structure. The tube wall is 0.05–0.10 mm thick. The new endosymbiotic tubeworm Coralloconchus bragensis n.
gen. and sp. shares zoological affinities with the tentaculitids (incertae sedis) and is assigned to the Family Cornulitidae (Tentaculita, Cornulitida).
endosymbiotic worms in the various invertebrates,
such as tabulate corals (Tapanila, 2004) and rugose corals
(Elias, 1986), are already common in the Ordovician (Tapanila,
2002, 2005), but they lack preserved mineralized shell. However,
if there were pre-Silurian endosymbiotic worm tubes of aragonitic
composition they would have been dissolved on the sea floor un-
less recrystallized early into low-Mg calcite (Cherns and Wright,
2000; Palmer and Wilson, 2004). Similarly, worm tubes of chi-
tinous composition would have disappeared. The exact biological
affinities of producers of worm-shaped Paleozoic bioclaustrations
are not known, but their morphology and ecology resemble mod-
ern polychaetes and vermetid gastropods (Tapanila, 2005). The
earliest known completely endosymbiotic animals with their own
exoskeleton are lingulid brachiopods from Heliolites corals in the
Silurian of Gotland (Richards and Dyson-Cobb, 1976), and from
the heliolitids, sarcinulid corals and stromatoporoids in the Silu-
rian of Anticosti Island (Tapanila and Copper, 2002; Tapanila and
Holmer, 2006). The ecology of the Ordovician gastropods en-
crusted by bryozoans (McNamara, 1978) and symbioses between
Silurian gastropod Semitubina sakoi Kase, 1986, and a favositid
coral (Kase, 1986) was different in that the host was a gastropod
on which bryozoan or coral larvae settled. In contrast to the Si-
lurian endosymbiotic lingulates, extensively encrusted Paleozoic
gastropods probably remained mobile and they definitely were not
endosymbiotic during their juvenile growth stage. The earliest
hitherto known endosymbiotic tubeworm-like fossils are Tor-
quaysalpinx Plusquellec, 1968 from the Middle Devonian stro-
matoporoids and tabulates (Stel, 1976; Zhen, 1996; Plusquellec,
1968; Tapanila, 2005) and Streptindytes Calvin, 1888 from the
Middle Devonian rugose corals (Calvin, 1888; Tapanila, 2005)
and stromatoporoids (Clarke, 1908; Tapanila, 2005). Darrel and
Taylor (1993) have discussed macrosymbioses in fossil corals.
Recently, all the Paleozoic endosymbiotic fossils have been
classified as ichnofossils (Tapanila, 2005), including those pos-
sessing their own skeleton such as Torquaysalpinx and Streptin-
dytes. However, the latter two do not meet the definition of an
ichnofossil (Bromley, 1970, 1996) because they contain fossilized
parts of the actual animal and thus are classical body fossils. Their
modern serpulid tubeworm analogues are similar, such as Spiro-
branchus giganteus (Pallas, 1766) in corals (Nishi and Nishihira,
1996) and Hydroides spongicola Benedict, 1887 in sponges,
which not only produce a trace (e.g. bioclaustration, see Palmer
and Wilson, 1988), but also secrete their own tube inside the host
skeleton. Their tubes are similar to those of substrate-cemented
and free-living serpulids (personal obs. O.V.). Several other Pa-
leozoic endosymbionts have previously been described as the
body fossils (Howell, 1962), such as Chaetosalpinx Sokolov,
1948 and Helicosalpinx Oekentorp, 1969. However, they lack
their own skeleton and are indeed bioclaustrations as interpreted
by Tapanila (2005).
In spite of striking morphological similarity to the modern ser-
pulid tubeworms (Weedon 1994; Taylor and Vinn, 2006), none of
the Paleozoic tubeworms such as sphenothallids (van Iten et al.,
1992; van Iten et al., 1996), cornulitids (Vinn and Mutvei, 2005),
microconchids (Weedon, 1990, 1991; Taylor and Vinn, 2006), try-
panoporids (Weedon, 1991) possess any of the key characters of
a serpulid tubeworm, such as the characteristic ultrastructure or
juvenile tube opened from both ends (Weedon, 1994; Vinn and
Mutvei, 2005; Taylor and Vinn, 2006). Most likely the earliest
true calcareous polychaete tubeworms appeared in the Mesozoic
(ten Hove and van den Hurk, 1993; Fischer et al., 2000; Vinn et
al., in press). Their characteristic tube structures and ultrastruc-
tures have been described from various Mesozoic worm fossils
(Senowbari-Daryan, 1997; Weedon, 1994; Vinn 2005a; Vinn et
al., in press).
The Silurian rocks (Wenlock-Pridoli) of Podolia (Ukraine) are
exposed in an approximately 80 km wide area along the river
Dniester and its tributaries (Fig. 1). The Silurian deposits were
formed in changing conditions from the normal-marine to lagoon
facies (Tsegelnjuk et al., 1983). The massive coral-stromatopo-
roid-algal bioherms (Grytsenko, 2007) from Lower to Upper Lud-
low of Podolia are characteristic of a shallow shelf environment.
The aim of this paper is to describe the earliest endosymbiotic
tubeworm and to discuss its biological affinities. This paper also
deals with the ecological and evolutionary implications of the
endosymbiosis of skeletal invertebrates.
The coral specimens with the endosymbiotic tubeworms occur
in the locality ‘‘Zhavanets 39,’’ which is located on the left bank
of the River Dniester between the villages of Zhvanets and Braga
(Fig. 1). The Grinchuk Member of the Rykhta Formation is ex-
posed above the water level of the river (Tsegelnjuk et al., 1983).
The Grinchuk Member corresponds to the Uncinatograptus cau-
datusMonograptus balticus Zone, upper Ludlow (Tsegelnjuk et
al., 1983) (Fig. 2). Heliolites sp. with Coralloconchus bragensis
n. gen. and sp. was collected from the layer of marl in the nodular
limestone approximately one meter above the water level from
the Grinchuk Member. The tabulate corals in the Grinchuk Mem-
ber at Zhavents 39 are represented by Squameofavosites incredi-
bilis (Tsegelnjuk et al., 1983), Syringoheliolites and at least five
species of Heliolites. Tesakov (1971) and Sokolov and Tesakov
(1984) described Favosites gothlandicus and Cystihalysites mir-
abilis from the Grinchuk Member near the village of Zhvanets.
We studied a collection of 185 tabulate corals from the Muksha
Member to the Grinchuk Member (Lower-Upper Ludlow). It in-
cludes seven species of Favosites, five species of Heliolites, one
species of Paleofavosites, 10 species of Cystihalysites, four spe-
cies of Cladopora, and two species of Syringopora.Riphaeolites,
Angopora,Stelliporella and Syringoheliolites are rare in collection
and are represented by only one species each. Material for study
1—Map of Podolia (Ukraine), showing the location of outcrop
Zhavanets 39 from which Coralloconchus bragensis n. gen. and sp. was col-
2— Stratigraphic context of the sample locality (after Tsegelnjuk
et al., 1983).
was collected from 13 localities. Two thin sections (one transverse
and one vertical section) were made of each coral specimen.
Thin sections containing tube worms were then polished and
etched with dilute (one per cent) acetic acid for 10 minutes and
coated with gold for SEM photography carried out with a Zeiss-
SEM at the Institute of Geology, University of Tartu.
Tubes of Recent calcareous annelids were selected for com-
parison with Coralloconchus bragensis n. gen. and sp. using the
database (unpublished, O.V.) of tube ultrastructures of 50 calci-
fying polychaete species. Examined Recent tubes of Glomerula
piloseta (Perkins, 1991) (Sabellidae), Pyrgopolon ctenactis
(Mo¨rch, 1863) (Serpulidae) and Dodecaceria coralii Leidi, 1855
(Cirratulidae), were embedded in epoxy resin, ground in longi-
tudinal and transverse direction, polished and etched with 1%
acetic acid for two minutes prior to SEM examination. Scanning
electron microscopy was carried out with a Hitachi S-4300 at the
Swedish Museum of Natural History, Stockholm.
All the specimens figured here are deposited at the Institute of
Geology, University of Technology, Tallinn (GIT 534-1-1, GIT
534-1-2, GIT 534-1-3, GIT 534-2-1).
Phylum I
Class T
Boucˇek, 1964
Order C
Boucˇek, 1964
Family C
Fisher, 1962
Genus C
new genus
Type species.Coralloconchus bragensis n. gen. and sp.
Diagnosis.A genus of Cornulitidae with small, slender, irreg-
ularly curved conical tubes with slowly increasing diameter.
Tubes have thin walls and a smooth lumen. Tube wall has a la-
mellar microstructure. Tubes are devoid of septa and vesicles in
the adult part and are not spirally coiled.
Other species.Monotypic thus far.
Etymology.Corallum Latin for ‘‘coral’’; concha Latin for ‘‘shell’’;
the shell of the new genus is embedded in coral.
new species
Figures 3, 4, 5.1–5.4
Description.Small, irregularly curved, slightly conical tubes with thin
walls (0.05–0.10 mm thick) embedded into host skeleton (coral) in their whole
length. Angle of divergence of tube about 10 degrees. Tube lumen smooth,
lacking septa at the tube diameters 0.5–1.05 mm. Exterior of tube mostly
smooth, rarely covered with variously developed perpendicular ridges. These
perpendicular ridges 50–60 m high, their inter val 0.1–0.3 mm. Shape of
ridges triangular in longitudinal section of tube. Tube wall up to twice as thick
at top of highest ridges as at interspaces between ridges. Tube wall lamellar;
lamellae 3.0–3.5 m thick, formed by calcite plates parallel to tube wall.
Types.Holotype GIT 534-1-1, and three paratypes (specimens GIT 534-
1-2, GIT 534-1-3, GIT 534-2-1) from the Silurian (Ludlow) of Zhavanets 39,
Podolia, Ukraine.
Etymology.After Braga village near the type locality of Zhavanets 39 in
Discussion.The irregularly curved tube, smooth tube lumen
and shape of the external perpendicular ridges of the new species
are similar to Conchicolites Nicholson, 1872 (Tentaculita, Cor-
nulitida, Cornulitidae) (Vinn and Mutvei, 2005), but it differs in
being embedded into the coral skeleton and in having lamellar
tube ultrastructure. The lamellar wall structure and lack of pseu-
dopuncta of the new species is similar to Septalites Vinn, 2005b
(Tentaculita, Cornulitida, Cornulitidae), but Coralloconchus bra-
gensis has much thinner walls and lacks the septa. The new spe-
cies is also somewhat similar to Trypanopora Sokolov and Obut,
1955 (Tentaculita, Trypanoporida) and Torquaysalpinx (Tentacu-
lita, Trypanoporida), but it differs in lacking vesicular wall struc-
ture and septa. The new species slightly resembles the endosym-
biotic tubeworm Streptindytes acervulariae Calvin, 1888
(Tapanila, 2005, p. 93, fig. 1.) from the Middle Devonian rugose
corals of Iowa, but C. bragensis differs in having an irregularly
curved tube instead of the regularly coiled, inverted cone-shaped
shell of Streptindytes. Unfortunately, the morphology of the ju-
venile growth stage of Coralloconchus bragensis is not known
and cannot be compared with the cornulitids, microconchids (Ten-
taculita, Microconchida) and trypanoporids (Tentaculita, Trypan-
oporida). The zoological affinities of Streptindytes need revision,
but most likely it is either a microconchid or trypanoporid tube-
Calcareous annelid tubes.In polychaete annelids, calcareous
tubes occur in serpulids (? Triassic-Recent) (ten Hove and van
den Hurk, 1993; Taylor and Vinn, 2006; Vinn et al., in press),
sabellids (Jurassic-Recent) (Ja¨ger, 2004; Vinn et al., in press), and
cirratulids (Oligocene—Recent) (Fischer et al., 2000). The fossil
and Recent calcareous annelid tubes possess a variety of ultra-
structures, but they are all different from that of Coralloconchus.
Their tubes usually are not lamellar (Fig. 5.6). Even in those
which are lamellar or contain lamellar layers, such as Recent
Glomerula piloseta (Sabellidae), Pyrgopolon ctenactis (Serpuli-
dae), and Dodecaceria coralii (Cirratulidae) (Fig. 5.5), the tube
lamellae have an aragonitic spherulitic prismatic ultrastructure
(Vinn et al., in press; Vinn, 2007) different from that of platy
homogeneous lamellae of Coralloconchus bragensis. Among cal-
careous polychaetes, only some Recent serpulids are thus far
known to be endosymbiotic (Nishi and Nishihira, 1996).
Vermiform gastropod shells.Some Recent and fossil gastro-
pods (vermetids and siliquariids) have worm-shaped shells with
31, 2, Coralloconchus bragensis n. gen. and sp. in Heliolites sp., Upper Ludlow, Silurian of Podolia, Ukraine, Scale bar 0.3 mm. 1, Longitudinal
section in plane-polarized light, paratype GIT 534-1-2; 2, cross section in plane-polarized light, holotype GIT 534-1-1, Scale bar 0.35 mm.
41–3, Coralloconchus bragensis n. gen. and sp. in Heliolites sp.,
Upper Ludlow, Silurian of Podolia, Ukraine, through the light microscope;
large black circle on the photo is pyritized Trypanites boring. 1, cross section,
holotype GIT 534-1-1; 2, oblique section, paratype GIT 534-1-3. 3, longitu-
dinal section, paratype GIT 534-1-2. The arrows point to the specimens. Scale
bar 2.5 mm.
the irregularly coiled free whorls. Their oldest fossils are known
from the Mesozoic (Savazzi, 1996; Bandel and Kowalke, 1997;
Bandel and Kiel, 2000; Rawlings et al., 2001). The siliquariids
are endosymbiotic and live in sponges (Savazzi, 1996). The tubes
of vermetid gastropods have three aragonitic layers (Wrigley,
1950) which differ from the single-layered lamellar calcite tube
wall of Coralloconchus, the siliquariid tube wall having crossed
lamellar structure (see Savazzi, 1996, p. 169, text-fig. 10 D). Their
tubes also have slits different from the complete tube wall of
Coralloconchus bragensis. Moreover, all unequivocal vermiform
gastropods are geologically too young (Bandel and Kowalke,
1997; Bandel and Kiel, 2000) to be candidates for Early Paleozoic
endosymbiotic tubeworms.
Tentaculitid tubeworms.A diverse fauna of tubeworms of
tentaculitid affinities is known from the Ordovician (Weedon,
1991; Weedon, 1994; Vinn, 2005b; Vinn and Mutvei, 2005; Tay-
lor and Vinn 2006; Vinn 2006a, 2006b; Vinn and Isakar, 2007).
Their diversity presumably reached a maximum by the Devonian.
The cornulitids range from the Ordovician (Caradoc) to the Car-
boniferous (Richards, 1974); trypanoporids are known only from
the Devonian (Weedon, 1991), microconchids range from the Or-
dovician (Ashgill) to Middle Jurassic (Taylor and Vinn, 2006;
Vinn and Taylor, 2007). The exact zoological affinities of tenta-
culitid tubeworms are not known, but most likely they represent
a group of extinct calcifying phoronids (Vinn, 2005b; Taylor and
Vinn, 2006). The earliest hitherto known endosymbiotic tenta-
culitid tubeworm is Torquaysalpinx from the Devonian. The ten-
taculitid tubeworms are characterized by lamellar tube structure
and closed, often bulbous, tube origins. Pseudopunctae are also
common in the tentaculitid tubeworms. The lamellar tube struc-
ture of C. bragensis closely resembles that of the other tentaculitid
tubeworms such as cornulitids, microconchids, trypanoporids.
The systematic position of C. bragensis within Tentaculita is also
well supported by the stratigraphic range of the new genus, which
coincides with the range of high diversity of the tentaculitid
Paleoecology of Coraliconchus.Coralloconchus bragensis is
found only in Heliolites sp. of the Silurian of Podolia. It is a
relatively rare endosymbiont in the tabulate corals and occurs in
only two Heliolites sp. specimens of 34 specimens studied. In
contrast, the traces of soft bodied worms, such as Chaetosalpinx
siberiensis are common in other corals and are much more abun-
dant. The restricted occurrence of C. bragensis in Heliolites sp.
presumably indicates host specificity of the tubeworm. Corallo-
conchus bragensis infested living hosts and caused slight changes
in the growth of the surrounding coral skeleton (Figs. 4.1). How-
ever, the tubeworm was probably not parasitic, because changes
in the coral skeleton are in the geometrical arrangement of coe-
nechymal tubuli around the worm tube rather than in the diameter
of the tubuli (Fig. 5.1). Among the Recent coral endosymbionts,
serpulid tubeworms could ecologically be most similar to C. bra-
Evolutionary implications.The calcareous exoskeletons of
probable ancestors of C. bragensis supported their soft tissues and
hence biological function. In endosymbiosis, the exoskeleton of
a tubeworm loses its original protective and supporting function.
This is different from soft bodied worms (corresponding trace
fossils e.g., Helicosalpinx;Chaetosalpinx ferganensis Sokolov,
1948; C. siberiensis Sokolov, 1948), which achieved their primary
protection against predators and supporting functions via adap-
tation to the endosymbiotic lifestyle. The Silurian (Wenlock–Lud-
low) cornulitids bear abundant shell repair marks (Vinn and Mut-
vei, 2005), which could be caused by predation. Similar shell
repairs occur in tubeworm-like Anticayltraea calyptrata (Eich-
wald, 1860) (Tentaculita) from the Silurian (Ludlow) of Baltos-
candia (Vinn and Isakar, in press). There are unambiguous ex-
amples of selective predation on a problematic colonial metazoan
51–4, Coralloconchus bragensis n. gen. and sp. in Heliolites sp., Upper Ludlow, Silurian of Podolia, Ukraine, SEM figures. 1, Cross section,
holotype GIT 534-1-1, showing the changes in the growth pattern of tubuli around the specimen; outer ring is penciled around the specimen for SEM
orientation; 2, close-up of the lower part of photograph 1, the arrows point to the external and internal surface of the tube wall of the specimen; 3, GIT 534-
1-1, close view of the cross section of the specimen’s tube wall, the arrows point to the external and internal surface of the tube wall of the specimen; 4,
GIT 534-1-1, detailed view of the lamellar tube wall of the specimen, showing the platy nature of the calcite lamellae forming the tube wall; 5, Dodecaceria
coralii Leidi, 1855, Recent, Mexico, cross section of the tube, showing the spherulitic prismatic ultrastructure of the aragonitic lamellae; 6, Pyrgopolon
ctenactis (Mo¨ rch, 1863), Recent, Bonaire, Netherlands Antilles, Caribbean; cross section of the outer nonlamellar layer of the tube, showing the semi-irregularly
oriented spherulitic prismatic crystals.
hederellid from the Devonian (Wilson and Taylor, 2006), and Ko-
walewski et al. (1998) have showed a significant increase in dril-
ling predation during the Devonian. The Devonian is also char-
acterized by the occurrence of two endosymbiotic tubeworm
genera, e.g., Torquaysalpinx and Streptindytes in tabulate corals,
rugose corals and stromatoporoids (Tapanila, 2005). One of pos-
sible reasons for the endosymbiosis of Coralloconchus could be
evolving boring predation capable of eliminating most of the pro-
tective advantages of the worm tube. The endosymbiotic C. bra-
gensis may also have been less endangered by overgrowth of the
other encrusting skeletons such as bryozoans as compared to en-
crusting cornulitids (probable ancestors).
We are grateful to J. Aruva¨li, Institute of Geology, University of Tartu, for
help with SEM and H. A. ten Hove, Zoological Museum, University of Am-
sterdam, for providing us with the identified tubes of Recent serpulids. Vinn
is grateful to H. Mutvei, Swedish Museum of Natural History, for help with
SEM and advice on the methodology for studying skeletal ultrastructures. We
are grateful to B. Pratt, L. Tapanila and anonymous reviewer for the useful
comments on the manuscript. V. Gritsenko is thanked for the help in collecting
material in Podolia. Vinn acknowledges financial support to projects NL-TAF-
111 and SE-TAF-1520 by SYNTHESYS, a program financed by the European
Commission under the Sixth Research and Technological Development
Framework Program ‘‘Structuring the European Research Area’’ and the Es-
tonian Science Foundation for the grant No. 6623 ‘‘Tube formation and bio-
mineralization in annelids, its evolutionary and ecological implications.’’ Mo˜-
tus research was supported by the Estonian target funding programme No.
0332524s03 and the Estonian Science Foundation grants Nos JD05-41 and
6127. M. Wilson read earlier versions of the Manuscript.
S. K
. 2000. Earliest known (Campanian) members of the
Vermetidae, Provannidae and Litiopidae (Cerithioidea, Gastropoda), and a
discussion of their possible relationships. Universita¨t Hamburg, Geologisch-
Pala¨ontologisches Institut, Mitteilungen, 84:209 –218
T. K
. 1997. Cretaceous Laxispira and a discussion
on the monophyly of vermetids and turritellids (Caenogastropoda, Mollus-
ca). Geologica et Paleontologica, 31:257–274.
, B. 1964. The Tentaculites of Bohemia. Publication of Czechoslo-
vakian Academy of Sciences, Prague, 125 p.
, R. G. 1970. Borings as trace fossils and Entobia cretacea Portlock,
as an example. In T. P. Crimes, and J. G. Harper, (eds.), Trace Fossils, 49–
90. Seel House Press, Liverpool.
, R. G. 1996. Trace fossils: Biology, taphonomy and applications.
(second edition). 361 pp. Chapman and Hall, London.
, S. 1888. On a new genus and new species of tubicolar Annelida.
American Geologist, 1:24–28.
V. P. W
. 2000. Missing mollusks as evidence of large-
scale, early skeletal aragonite dissolution in a Silurian Sea. Geology, 28:
, J. M. 1908. The beginning of dependent life. New York State Mu-
seum Bulletin, 121:146–196.
P. D. T
. 1993. Macrosymbiosis in corals: A review
of fossil and potentially fossilizable examples. Courier Forschunginstitut
Senckenberg, 164:185–198.
, E. 1860. Lethaea Rossica ou Pale´ontologie de la Russie, de´ crite
et figure´e. 1, Seconde Section de l’ancienne Pe´ riode. Stuttgart, 1104–1105.
, R. J. 1986. Symbiotic relationships between worms and solitary corals
in the Late Ordovician. Paleobiology, 12:32–45.
, D. W. 1962. Small conoidal shells of uncertain affinities, p. 130–143.
In R. C. Moore (ed.), Treatise on Invertebrate Paleontology, Pt. W, Geo-
logical Society of America and University of Kansas Press, Lawrence, Kan-
, R., B. P
J. R
. 2000. Organomineralization of
Cirratulid Annelid Tubes-Fossil and Recent Examples. Facies, 42:35–50.
, V. 2007. Distribution of corals on the Silurian Podolian Shelf,
p. 185–198. In B. Hubmann and W. E. Piller (eds.), Fossil Corals and
Sponges. Proceedings of the 9th International Symposium on Fossil Cni-
daria and Porifera, Graz 2003. O
¨sterreichische Akademie der Wissenschaf-
ten Schriftenreiche der Erdwissenschaftlichen Kommissionen, 17.
, 1993. A review of Recent and fossil
serpulid ‘‘reefs’’; actuopaleontology and the ‘‘Upper Malm’’ serpulid lime-
stones in NW Germany. Geologie en Mijnbouw, 72:23–67
, B. F. 1962. Worms. In R. C. Moore (ed.), Treatise on Invertebrate
Paleontology, Pt. W, Miscellanea: Conodonts, conoidal shells of uncertain
affinities, worms, trace fossils and problematica, F144–F177. Geological
Society of America, Boulder, Colorado, and University of Kansas, Lawr-
ence, Kansas.
R. H. M
. 1992. New data on the mor-
phology of Sphenothallus Hall: Implications for its affinities. Lethaia, 25:
R. S. C
. 1996. Problematical fossil cni-
darians from the upper Ordovician of the north-central USA. Palaeontology,
, M. 2004. Serpulidae und Spirorbidae (Polychaeta sedentaria) aus Cam-
pan und Maastricht von Norddeutschland, den Niederlanden, Belgien und
angrenzenden Gebieten. Geologisches Jahrbuch, A157:121–249.
, T. 1986. Mode of life of the Silurian uncoiled gastropod Semitubina
sakoi n.sp. from Japan. Lethaia, 19:327–337.
, M., A. D
F. T. F
. 1998. A fossil record full
of holes, the Phanerozoic history of drilling predation. Geology, 26:1091–
, K. J. 1978. Symbiosis between gastropods and bryozoans in the
late Ordovician of Cumbria, England. Lethaia, 11:25–40.
, O. A. L. 1863. Revisio Critica Serpulidarum. Et bidrag til rørormenes
naturhistorie. Naturhistorisk Tidskrift, Henrik Krøyer, København, 1(3):
, H. A. 1872. On the genera Cornulites and Tentaculites and a
new genus Conchicolites. American Journal of Science, 3:202–206.
M. N
. 1996. Age-estimation of the Christmas Tree
worm Spirobranchus giganteus (Polychaeta, Serpulidae) living in the coral
skeleton from the coral-growth band of the host coral. Fisheries Sciences,
, K. 1969. Kommensalismus bei Favositiden. Mu¨nstersche For-
schungen zur Geologie und Pala¨ ontologie, 12:165–217.
M. A. W
. 1988. Parasitism of Ordovician bryozoans
and the origin of pseudoborings. Palaeontology, 31:939–949.
M. A. W
. 2004. Calcite precipitation and dissolution
of biogenic aragonite in shallow Ordovician calcite seas. Lethaia, 37:417–
, T. H. 1991. Calcisabella piloseta, a new genus and species of Sa-
bellinae (Polychaeta: Sabellidae). Bulletin of Marine Science, 48:261–267.
, Y. 1968. Commensaux des tabule´s et stromatoporoı¨des du De´-
vonien armoricain. Annales de la socie´te´ge´ologique du Nord, 88:47–56.
, T. A., T. M. C
R. B
. 2001. A Major Mitochon-
drial Gene Rearrangement Among Closely Related Species. Molecular Bi-
ology and Evolution, 18(8):1604–1609.
, P. R. 1974. Ecology of the Cornulitidae. Journal of Paleontology,
M. D
. 1976. A Lingula-Heliolites associa-
tion from the Silurian of Gotland, Sweden. Journal of Paleontology, 50:
, E. 1996. Adaptations of vermetid and siliquariid gastropods. Pa-
laeontology, 39:157–177.
, B. 1997. Barbafera carnica Senowbari-Daryan, 1980:
A Triassic Worm-tube. Facies, 36:57–68.
, B. S. 1948. Kommenzializm u Favositid. Izvestija Akademii Nauk
SSSR, Biology Series, 1:101–110. (In Russian).
A. O
. 1955. In B. S. Sokolov (ed.), Paleozoic tab-
ulates from the European part of the USSR. Trudy VNIGRI, NS, 85. 527
pp. (In Russian).
J. I. T
. 1984. Populyatsionnyj, biotsenoticheskij
i biostratigraficheskij analiz tabulyat. Podolskaya model. Akademiya Nauk
SSSR, Sibirskoe Otdelenie, Institut Geologii i geofiziki, 577, 198 pp. (In
, J. H. 1976, The Paleozoic hard substrate trace fossils Helicosalpinx,
Chaetosalpinx and Torquasalpinx. Neues Jahrbuch fu¨r Geologie und Pa-
la¨ontologie, 12:726–744.
, L. 2002. A new endosymbiont in Late Ordovician tabulate corals
from Anticosti Island, eastern Canada. Ichnos, 9:109–116.
, L. 2004. The earliest Helicosalpinx from Canada and the global
expansion of commensalism in the Late Ordovician sarcinulid corals (Ta-
bulata). Palaeogeography, Palaeoclimatology, Palaeoecology, 215:99–110.
, L. 2005. Palaeoecology and diversity of endosymbionts in Palaeo-
zoic marine invertebrates: Trace fossil evidence. Lethaia, 38:89–99.
P. C
. 2002. Endolithic trace fossils in Ordovician-
Silurian corals and stromatoporoids, Anticosti Island, eastern Canada. Acta
Geologica Hispanica, 37:15–20.
L. E. H
. 2006. Endosymbiosis in Ordovician-Silu-
rian corals and stromatoporoids: A new lingulid and its trace from eastern
Canada. Journal of Paleontology, 80:750–759.
O. V
. 2006. Convergent morphology in small spiral
worm tubes (‘‘Spirorbis’) and its palaeoenvironmental implications. Jour-
nal of the Geological Society, London, 163:225–228.
, J. I. 1971. Favositidy Podolii. Akademiya Nauk SSSR, Sibirskoe
Otdelenie, Trudy Instituta Geologii i Geofiziki, 120 pp. (In Russian).
, P. D., V. P. G
S. Z
V. N. S
1983. Silur Podolii, Putevoditel ekskursii. 244 pp., Naukova Dumka, Kyiv.
, O. 2005a. The tube ultrastructure of serpulids (Annelida, Polychaeta)
Pentaditrupa subtorquata, Cretaceous, and Nogrobs cf. vertebralis, Juras-
sic, from Germany. Proceedings of the Estonian Academy of Sciences,
Geology, 54:260–265.
, O. 2005b. A new cornulitid genus from the Silurian of Gotland, Swe-
den. GFF, 127,205–210.
, O. 2006a. Tentaculitoid affinities of the tubeworm-like fossil Tymbo-
choos sinclairi (Okulitch, 1937) from the Ordovician of North America.
Geobios, 39:739–742.
, O. 2006b. Two new microconchid (Tentaculita Boucˇek 1964) genera
from the Early Palaeozoic of Baltoscandia and England. Neues Jahrbuch
fu¨ r Geologie und Pala¨ontologie, 2006(2):89–100.
, O. 2007. Taxonomic implications and fossilization of tube ultrastructure
of some Cenozoic serpulids (Annelida, Polychaeta) from Europe. Neues
Jahrbuch fu¨ r Geologie und Pala¨ ontologie, 244(1):115–128.
H. M
. 2005. Observations on the morphology, and affin-
ities of cornulitids from the Ordovician of Anticosti Island and the Silurian
of Gotland. Journal of Paleontology, 79:725–736.
M. I
. 2007. Tentaculitid affinities of Anticalyptraea from
the Silurian of Baltoscandia. Palaeontology, 50:1385–1390.
P. D. T
. 2007. Microconchid tubeworms from the Ju-
rassic of England and France. Acta Palaeontologica Polonica, 152:391–399.
, O., H. A.
H. M
. In press. On the tube ultrastruc-
ture and origin of calcification in sabellids (Annelida, Polychaeta). Palaeon-
, M. J. 1990. Shell structure and affinity of vermiform ‘‘gastropods’’.
Lethaia, 23:297–309.
, M. J. 1991. Microstructure and affinity of the enigmatic Devonian
tubular fossil Trypanopora. Lethaia, 24:227–23.
, M. J. 1994. Tube microstructure of Recent and Jurassic serpulid
polychaetes and the question of the Palaeozoic ‘‘spirorbids’’. Acta Palaeon-
tologica Polonica, 39:1–15.
P. D. T
. 2006. Predatory drillholes and partial
mortality in Devonian colonial metazoans. Geology, 34:565–568.
, A. 1950. The differences between the calcareous tubes of vermetids
and of serpulids. Journal de Conchyliologie, 90:118–121.
, Y.-Y. 1996. Succession of coral associations during a Givetian trans-
gressive-regressive cycle in Queensland. Acta Palaeontologica Polonica, 41:
13 J
... Herringshaw et al. (2007) reviewed previous proposals and hypothesized alternatively that the cornulitids represent solitary, aseptate stem-zoantharians (Cnidaria: Anthozoa). Despite continuing uncertainty about origins and relationships, understanding of these problematic fossils continues to evolve, for example with the recognition of different skeletal ultrastructures in different cornulitid genera (Vinn and Mutvei, 2005), and the discovery of commensal Silurian cornulitids within compound corals and stromatoporoids (Vinn and Mö tus, 2008;Vinn and Wilson, in press). Fisher (1962) included the four genera Cornulites Schlotheim, 1820, Conchicolites Nicholson, 1872a, Cornulitella Nicholson, 1872b, and Kolihaia Prantl, 1944 in the Family Cornulitidae. ...
... The studied specimens have been referred to Cornulites in view of the principal characteristics in common with the genus, and it is proposed that the diagnosis be emended to include forms having shell walls with a distinctive outer layer consistently preserved as prismatic calcite. The endobiotic Coralloconchus differs from Cornulites in having a non-cellular, microlamellar shell wall (Vinn and Mö tus, 2008). The shell wall structure of other cornulitid genera was not reported, precluding further comparison. ...
... Great diversification among nonskeletal endobionts in the Late Ordovician (e.g., Tapanila, 2004Tapanila, , 2005 was attributed by Huntley and Kowalewski (2007) to increased predation pressure. Vinn and Mö tus (2008) observed that Silurian epibiotic cornulitids bear abundant shell markings suggestive of predation and speculated that evolution of an endobiotic life style in the cornulitids may also have been a consequence of predation pressure. The restricted occurrence of C. celatus contrasts markedly with the widespread occurrence of the Early Silurian endobiotic species C. stromatoporoides in Estonia (Vinn and Wilson, in press). ...
Conoidal shells of Cornulites celatus n. sp. occur commonly within host coralla of Propora conferta Milne-Edwards and Haime, 1851, sensu lato, from the Laframboise Member of the Ellis Bay Formation (Ashgill: Upper Ordovician) at Pointe Laframboise on western Anticosti Island. Examples have also been found at the same locality in the tabulate corals Paleofavosites sp., Acidolites arctatus Dixon, 1986, and A. compactus Dixon, 1986, and the stromatoporoid Ecclimadictyon sp., but not in other associated tabulate coral species. Growth interference between the shells and their hosts indicates a commensal relationship. C. celatus apparently had a more limited paleoenvironmental range than its principal coral host species, which occurs abundantly elsewhere on the island without its endobiotic partner. The diagnosis of Cornulites is emended to include forms having a two-layered shell wall with a distinctive outer layer consistently preserved as prismatic calcite. This new species extends the known stratigraphic range of cornulitids in commensal relationships with corals and stromatoporoids from the Silurian back to the Upper Ordovician.
... Secondarily free-living forms occurred among problematic tubeworms (e.g., Cornulites [37], probably also tentaculitids [28]) and among calcareous polychaetes (e.g., Ditrupa [38]; Rotularia, Glomerula [39]). Endosymbiotic forms evolved among problematic tubeworms [28], e.g., in Silurian cornulitids [40], Devonian trypanoporids [41], and Jurassic to Recent polychaete serpulids [30]. ...
... Reef-forming gregarious forms occurred among prob-lematic tubeworms [40], e.g., in the Ordovician Tymbochoos, Devonian to Carboniferous microconchids [42], Upper Triassic to Recent polychaete serpulids [5], and Oligocene to Recent polychaete cirratulids [25]. Freshwater environments were colonized from the Carboniferous to Triassic by microconchids [32], which were widely distributed and more successful in colonization of freshwater than calcareous polychaetes [40]. ...
... Reef-forming gregarious forms occurred among prob-lematic tubeworms [40], e.g., in the Ordovician Tymbochoos, Devonian to Carboniferous microconchids [42], Upper Triassic to Recent polychaete serpulids [5], and Oligocene to Recent polychaete cirratulids [25]. Freshwater environments were colonized from the Carboniferous to Triassic by microconchids [32], which were widely distributed and more successful in colonization of freshwater than calcareous polychaetes [40]. The only Recent species of freshwater calcareous polychaetes, the serpulid Marifugia cavatina, lives in caves in Herzegovina [30]. ...
Full-text available
Polychaete serpulids are globally distributed tubeworms mostly in marine environments from Late Triassic to modern time. These calcareous tubeworms could be rock-forming, reef-building, or a principal fouling organism in harbor and bays. Carbonates of the Paleogene Kalatar Formation in southwest Tarim Basin yield abundant serpulid fossils, which, together with oyster fossils, constitute the characteristic fossil assemblage of the Kalatar Formation. Other common fossils include bivalves, gastropods, ostracods, echinoderms, and bryozoans. Lithologies that yielded serpulid fossils are characterized by micritic bioclastic limestone, sandy limestone, and shelly limestone, indicating a semi-restricted to open shallow marine environment with medium to low water energy. The research data about serpulids and their fossil materials from China are relatively rare. Based on the studies of fossils taxonomy, community palaeoecology, and fossil taphonomy, this paper analyzed and studied the types, occurrence, distribution, and morphological characteristics of serpulids and their palaeoecological features in the Kalatar Formation. Two serpulid community compositions were recognized in the Kalatar Formation, including a rock-forming Ditrupa community and a cluster-growth Propomatoceros community. The Ditrupa community was distributed in coastal environment of the west Kunlun piedmont, lived on sandy hard substrates with little mud, and rarely occurred in lagoon and tidal settings. The Propomatoceros community occurred in offshore middle carbonate ramp in the piedmont of the south Tianshan Mountains and in offshore shelf in the piedmont of the west Kunlun Mountains. According to the analysis on the host-rock lithologies, preservation and symbionts, it is inferred that serpulids in the Kalatar Formation grew on the oyster shell or other hard substrate, and they did not form reefs or bioherms.
... Over the previous decades, tentaculitoids have been referred to several invertebrate groups and classified as the spines of brachiopods (Towe 1978), as cephalopods bereft of a siphuncle, as pteropod-like pelagic gastropods (Orlov et al. 1976;Weedon 1990; Drapatz 2010), as a distinct class of Mollusca (Wittmer and Miler 2011;Fregatto and Vega 2015; references therein), oras lophophorates (Weedon 1990(Weedon , 1991(Weedon , 1994Mutvei 2005, 2009;Vinn 2005aVinn , 2006aVinn and Isakar 2007;Vinn and Mõtus 2008;Taylor et al. 2010) There are currently two distinct ideas about the phylogeny of tentaculitoids: (і) tentaculitoids is a distinct class of phylum Mollusca owing to the similarity in the wall structure and shell morphology (Wittmer and Miler 2011;Fregatto and Vega 2015;Comniskey and Ghilardi 2018; references therein) that is still questioned (Wittmer and Miler 2011); (іі) tentaculitoids are closely related to cornulitids and microconchids (Fregatto and Vega 2015;Comniskey and Ghilardi 2018;references therein). In the preceding decade, several researchers interpreted the phylogeny of tentaculitoids (Vinn and Mutvei 2009;Taylor et al. 2010) and also their evolutionary palaeoecology (Vinn 2010a). ...
... The generic heterogeneity trend of tentaculitoid tubeworms conforms to the prevailing diversity trend of the Palaeozoic period's evolutionary fauna (Stanley 2007). They have the highest morphological diversity in the Middle Palaeozoic (Bouček 1964;Richards 1974;Larsson 1979;Vinn 2004Vinn , 2005Vinn 2006a;Taylor and Wilson 2008;Vinn and Mõtus 2008;Vinn and Wilson 2010;Vinn et al. 2014Vinn et al. , 2019Vinn and Toom 2020;Vinn et al. 2021), followed by a gradual decrease in the late Palaeozoic (Vinn and Mutvei 2009;Vinn 2010a). The reduction in the morphological and ecological diversity of tentaculitoid tubeworms in the late Palaeozoic time was possibly concomitant to the diminution of the tropical environments and the decline of prevailing ecological niches during the icehouse climate in the late Palaeozoic (Cocks 2007;Angiolini et al. 2009). ...
Tentaculitoid tubeworms, an extinct invertebrate group, is being reported from the Ordovician strata of Spiti, Tethyan Himalaya, India. External and internal moulds of tentaculitoids are preserved in the lower unit of the Takche Formation which is considered early Late Ordovician in age. The assemblage includes Cornulites cf. sterlingensis, Cornulites zatoni, Cornulites sp., and Tentaculites spp. Cornulites cf. sterlingensis, Cornulites sp., and Cornulites zatoni are recorded and described for the first time from the Ordovician strata of India. Though, the occurrence of genus Tentaculites is previously mentioned from the Early Palaeozoic sequence of Spiti, the detailed documentation is provided herein for the first time from the Ordovician sequence of Spiti, India. The occurrence of solitary-free forms of tentaculitoids from shallow marine settings of Spiti, Tethyan Himalaya, India indicates that unattached free forms might have been more competent in the siliciclastic environment of relatively low palaeolatitude than carbonate platforms of lower palaeolatitudes and silici-clastic environments of the highest palaeolatitudes. ARTICLE HISTORY
... Among tabulates, only associations with possible worms (i.e. Chaetosalpinx), cornulitids and lingulids are known (Richards & Dyson Cobb 1976;Vinn & Mõtus 2008;Mõtus & Vinn 2009). Surprisingly, cases of symbiosis with bryozoans seem to have been rare. ...
... Unlike true tabulate corals, the heliolitids had significant coenosteal skeleton between corallites; this spongy skeletal tissue may have been ideal for embedment of sclerobionts while preventing damage to the actual coral polyps. The symbionts' influence on the host Heliolites is difficult to determine on the fossil material (Vinn & Mõtus 2008). ...
Full-text available
Thirteen symbiotic associations occur in the Silurian of Baltica. Symbiosis was especially prominent among colonial animals, most commonly with stromatoporoids. These sponges hosted the most diverse fauna of endobiotic symbionts (including rugosans, Syringopora, ‘polychaetes’, cornulitids and lingulids). This pattern can be explained by the abundance of stromatoporoids in the Silurian of Baltica and their large skeletal volume, making them attractive hosts for smaller invertebrates. There is an evolutionary trend of an increasing number of different pairs of symbiotic taxa from the Llandovery to the Ludlow, with a remarkable increase in the Ludlow. This is likely related to an increase in the number of mutualistic taxa that could have had evolutionary advantages over organisms less amenable to symbiosis. The number of different pairs of symbiotic taxa also increased in the Wenlock, which may be linked to delayed recovery from the end-Ordovician mass extinction.
... 此外, 有疑问的管状蠕虫(如 Cornulites, 见 Richards [37] [38] ; Rotularia, Glomerula [39] )中也有一部分为 自由生活形式. 内共生的生态类型以一些有疑问的 管 状 蠕 虫 , 如 志 留 纪 的 Cornulitids [40] , 泥 盆 纪 的 trypanoporids [41] 和侏罗纪到现代的多毛类龙介类 [28] 为代表. 造礁群居生态类型以一些有疑问的管状蠕 虫 , 如 奥 陶纪 的 Tymbochoos, 泥 盆 纪 到 石 炭纪的 microconchids [42] , 晚三叠纪到现代的多毛类龙介虫 [5] 和渐新世到现代的多毛类 Cirratulids [25] [30] . ...
... The hitherto earliest known cornulitid endobiotic symbionts are found in Hirnantian tabulates of North America (Dixon, 2010). Prior to this work the earliest known endobiotic cornulitids from Baltica were Silurian (Sheinwoodian) Cornulites stromatoporoides from Estonia and Coralloconchus bragensis from the Ludlow of Podolia (Ukraine) (Vinn and Mõtus, 2008). ...
A diverse early endobiotic coral symbiont assemblage has been detected within heliolitid tabulate corals of Katian age from northwestern Estonia. This assemblage indicates that the earliest endobiotic coral symbiont communities were not restricted to North America. The symbiotic endobiont assemblage comprises abundant Cornulites aff. celatus and rare Conchicolites hosholmensis and Chaetosalpinx sp. The cornulitids are the earliest known symbiotic endobionts of this group. The symbiotic endobionts presumably occurred in certain hosts only and preferred Protoheliolites dubius over Propora speciosa. This record indicates that early endobiotic cornulitids were more diverse than previously thought and appeared in large numbers in the Katian. It is possible that the appearance of abundant skeletal symbiotic coral endobionts may coincide with the Global Ordovician Biodiversification Event, supporting the hypothesis of escape from increased predation.
Colonial species occur in a wide range of aquatic invertebrates, some having excellent fossil records, notably corals, bryozoans and graptolite hemichordates. In contrast to unitary animals, colonial animals grow by adding repetitive modules known as zooids. The ability of colonies to endure partial mortality and the typically plastic growth of benthic colonial species facilitates the formation of macrosymbiotic associations, some of which may be parasitic. However, as with unitary fossils, it is notoriously difficult to identify whether the symbioses are parasitisms (+/-) or mutualisms (+/+). Intergrowths between host colonies of stromatoporoid sponges, corals or bryozoans, and skeletal or soft-bodied symbionts are particularly common in Ordovician-Devonian shallow-water deposits. Soft-bodied symbionts in such intergrowths are represented by moulds in the host skeletons, a process of preservation termed bioclaustration. As yet, however, there is a lack of convincing data showing that any of these symbionts were parasites. By comparison with modern analogues, some fossil galls provide more convincing examples of parasitism, and the destructive effects of borings into the skeletons of benthic colonies also argue in favour of parasitism. Pelagic graptoloid hemichordates from the Early Palaeozoic occasionally contain cysts or tubes that have been attributed to parasites on the grounds that they would have adversely affected the hydrodynamics of the floating colonies. Future studies should test for parasitism by comparing the sizes of colonies hosting symbionts with those lacking symbionts.
Full-text available
Symbiotic associations are a poorly studied aspect of the fossil record, owing largely to the taphonomic biases that inhibit direct observation that two organisms shared an intimate association in life. A symbiosis between an infesting animal and a skeleton-producing host can form a bioclaustration cavity that directly preserves the association and has a high preservation potential. Identifi cation of ancient mutuals and parasites must reject the null hypothesis of commensalism by demonstrating that the symbiosis correlates with a positive or negative change in host fi tness as compared to a non-symbiotic relative of the host taxon. Reviews of the Paleozoic record of marine symbionts show that the majority are hosted by colonial animals, especially corals and calcareous sponges. These hosts include structural forms that have moderate to high levels of integration and can support bioclaustrations between clonal units, mitigating the negative effects of symbionts, and perhaps facilitating the symbiosis. The fossil record is biased toward recording long-lasting, widespread, equilibrated associations. By contrast, parasitisms that are especially negative to the host are expected to be fossilized rarely. The symbiotic associations that form bioclaustrations may also represent an endolithic adaptive strategy in response to biological antagonisms, such as predation and spatial competition. The Late Ordovician rise in symbiotic bioclaustrations joins burrows and borings as trace fossil examples of infaunalization strategies that accompany the Ordovician faunal radiation.
Full-text available
Two species of endobiotic cornulitids are described from heliolitid corals of the Katian (Late Ordovician) of Estonia. They are thus far the earliest known showing this symbiotic relationship. Conchicolites hosholmenis n. sp. represents the only known symbiotic coral endobiont species in the genus. Cornulites sp. aff. Cornulitescelatus closely resembles the North American endobiotic species C. celatus from tabulates of the Hirnantian (Late Ordovician).
Full-text available
Tentaculitoids have traditionally been assigned to either Mollusca or Annelida. In addition, cnidarian, brachiopod, bryozoan, phoronid and sponge affinities have been proposed. Similarity analyses carried out with tentaculitoid characters place them in a cluster together with Brachiozoa (Brachiopoda+Phoronida). Tentaculitoids are less similar to Mollusca and Bryozoa. No support was found for the annelid, cnidarian and sponge affinities. Thus, tentaculitoids belong to Lophotrochozoa with a high degree of certainty. They are best classified as "lophophorates", most likely Brachiozoa.
Sphenothallus sp. and eight species of conulariids, distributed among the genera Climacoconus, Conularia, Glyptoconularia and Metaconularia, occur in the Elgin Member of the Maquoketa Formation (upper Ordovician) of north-eastern Iowa and south-eastern Minnesota, USA. Seven of the eight conulariid species exhibit internal test structures at their corners and/or midlines. Comparisons of these test structures with internal thecal structures of coronatid scyphozoans corroborate the hypothesis that conulariids were more closely related to scyphozoan cnidarians than they were to any other extant taxon of comparable rank. Sphenothallus and conulariids occur in all four Elgin Member biofacies. However, the distribution of Climacoconus and Conularia is facies-dependent, with Climacoconus occurring predominantly in the brachiopod-echinoderm biofacies and the basal Maquoketa phosphorite, Conularia splendida predominantly in the trilobite-dominated biofacies, and C. trentonensis predominantly in the brachiopod-echinoderm, mixed faunas and graptolite shales biofacies. Conulariids commonly occur in monospecific clusters, possibly clonal in origin, and some specimens show orientational evidence of original attachment to Sphenothallus or nautiloid shell material. Together with previously reported data on the distribution and biostratinomy of Sphenothallus and conulariids, these results suggest that both taxa were sessile benthic organisms that inhabited all major Elgin Member bottom environments, including a shallow, oxic carbonate shelf and a deeper, dysoxic shelf margin and shale basin slope. One new species, Climacoconus sinclairi, is described.
Symbiotic relationships involving physical contact between worms and solitary rugosan polyps are recorded by the following structures in North American Late Ordovician corals: (1) Trypanites borings enclosed within septal swellings in two specimens, (2) vermiform grooves and openings along the external wall of one corallum, and (3) a chamber containing a unique brown tube within one individual. These features are indicative, respectively, of commensal boring polychaete annelids that penetrated through coralla, commensal epizoic worms of unknown taxonomic affinity that attached to the side of a polyp, and a tubicolous worm (possibly a polychaete) that was likely a parasitic endozoan. Symbionts comparable to the latter two types are also known from two specimens of Devonian solitary rugose corals. Indirect evidence suggests that symbioses between solitary rugosans and the worms that produced Trypanites borings as dwelling structures in the sides of coralla were relatively common. However, direct evidence that the hosts were alive has been found in only two corals. In both cases, worms bored through septa within the calices and came into contact with basal surfaces of the polyps, which secreted skeletal material that sealed off the intruders. The rarity of such structures suggests that the encounters were inadvertent. If boring worms favored upcurrent portions of objects in order to maximize feeding benefits and avoid sedimentation, their locations indicate that the concave sides of curved coralla faced toward prevailing currents when in life positions. “Opportunistic” worms are known to have attached to the sides of polyps only in rare instances when the hosts became temporarily exposed as a result of accidents or abnormalities. This indicates that coralla normally served to shield polyps from colonization by nonboring epizoans. Worms that apparently extended up through openings in basal surfaces of polyps likely obtained sustenance parasitically within the central cavities. They could have entered the hosts through their mouths, or via the calices when parts of the polyps detached from their coralla and contracted radially. The rarity of this type of relationship in solitary Rugosa suggests that the worms entered inadvertently. Symbioses involving physical contact between worms and polyps seem to have been rare throughout the history of solitary rugose corals. Both groups apparently tolerated such associations when they did occur, although the rugosans secreted structures in their coralla that served to isolate the symbionts. In doing so, they recorded the presence of worms not likely to be preserved as body fossils. The interpretation of such features provides information on the physiology and ethology of both organisms, on the history of symbiotic relationships, and on the diversity of soft-bodied organisms in ancient environments.