taphonomy and affinity of an enigmatic silurian vertebrate, jamoytius ...€¦ · taphonomy and...
TRANSCRIPT
TAPHONOMY AND AFFINITY OF AN ENIGMATIC
SILURIAN VERTEBRATE, JAMOYTIUS KERWOODI
WHITE
by ROBERT S. SANSOM* , KIM FREEDMAN*� , SARAH E. GABBOTT* ,
RICHARD J. ALDRIDGE* and MARK A. PURNELL*�*Department of Geology, University of Leicester, University Road, Leicester LE1 7RH, UK; e-mails [email protected], [email protected], [email protected],
�6 Wescott Road, Wokingham, Berkshire RG40 2ES, UK; e-mail [email protected]
�Author for correspondence
Typescript received 14 July 2009; accepted in revised form 23 April 2010
Abstract: The anatomy and affinities of Jamoytius kerwoodi
White have long been controversial, because its complex
taphonomy makes unequivocal interpretation impossible
with the methodology used in previous studies. Topological
analysis, model reconstruction and elemental analysis, fol-
lowed by anatomical interpretation, allow features to be
identified more rigorously and support the hypothesis that
Jamoytius is a jawless vertebrate. The preserved features of
Jamoytius include W-shaped phosphatic scales, 10 or more
pairs of branchial openings, optic capsules, a circular, subter-
minal mouth and a single terminal nasal opening. Interpreta-
tions of paired ‘appendages’ remain equivocal. Phylogenetic
analysis places Jamoytius and Euphanerops together (Jamoytii-
formes), as stem-gnathostomes rather than lamprey related
or sister taxon to Anaspida.
Key words: Jamoytius, Euphanerops, phylogeny, taphonomy,
Vertebrata, Gnathostomata, Silurian.
W hite (1946) first described Jamoytius kerwoodi on the
basis of two specimens from a Lower Silurian (Llando-
very) horizon of the Lesmahagow inlier of Lanarkshire,
Scotland, and considered it to be the most primitive
known vertebrate. Numerous subsequent authors have
disputed this conclusion or have disagreed with aspects of
White’s (1946) interpretation (see Supporting Informa-
tion, Data S1). Evidence from additional specimens col-
lected at Lesmahagow localities (see Ritchie 1968, 1985
for synopses of locality and stratigraphy details) prompted
Ritchie (1960, 1963, 1968, 1984) to redescribe Jamoytius
as an ‘unspecialized anaspid’, possibly related to the
extant jawless vertebrates. As shown by Plate 1, most of
Ritchie’s (1960, 1968) interpretations of the anatomical
features of Jamoytius differ from those of White (1946).
Despite Ritchie’s treatment, Jamoytius continued to
generate debate. Various authors challenged Ritchie’s
work (e.g. Janvier 1981; Forey and Gardiner 1981) or sug-
gested affinities with a range of subsequently discovered
fossils (e.g. Janvier and Busch 1984; Briggs and Clarkson
1987). Cladistic analyses of the jawless vertebrates (e.g.
Janvier 1981, 1996a, b; Forey 1984, 1995; Forey and Jan-
vier 1993) have also failed to clarify the affinities of
Jamoytius; the lack of agreement regarding its anatomical
homologies has meant that different analyses have used
different character codings. In addition to coding, choice
of the in-group taxa included in the phylogenetic investi-
gation has affected the placement of Jamoytius (Donoghue
et al. 2000; Donoghue and Smith 2001; Gess et al. 2006).
The position of Jamoytius on cladograms has conse-
quently not stabilized, though Jamoytius usually appears
as a sister taxon to the lampreys, the anaspids, or Eupha-
nerops longaevus Woodward, 1900.
Janvier and Lund (1983, p. 412) wrote, ‘These various
interpretations cause one to wonder about the degree of
imagination involved in the study of Jamoytius and other
fossils preserved as tarry impressions.’ So why has it
proved so difficult to produce a definitive interpretation
of Jamoytius and to determine its affinities? The principal
problem is one that affects interpretation of many prob-
lematic fossils – disentangling the different aspects of the
process of anatomical reconstruction. The selection of an
appropriate anatomical comparator upon which to base
hypotheses of homology (comparisons being drawn either
directly to a specific extant organism or clade, or indi-
rectly through a fossil intermediate) can be especially
problematic (Donoghue and Purnell 2009); the choice of
interpretative model needs careful justification on the
basis of characters present, preferably unequivocally, in
the fossils. In the case of Jamoytius, almost all workers
have considered it as a jawless vertebrate without explicit
justification, and consequently, anatomical interpretations,
[Palaeontology, Vol. 53, Part 6, 2010, pp. 1393–1409]
ª The Palaeontological Association doi: 10.1111/j.1475-4983.2010.01019.x 1393
homology statements and phylogenetic analyses are all at
risk of circularity. Jamoytius is generally preserved in two
dimensions, and investigations have often concentrated on
relating the shape of its features to the three-dimensional
anatomical parts of presumably related organisms. Differ-
ent workers have alternatively described the same circular
features in Jamoytius as mouth, eyes and nasal structures
(Pl. 1 and Supporting Information, Data S1). These inter-
pretations are thus inherently equivocal. Anatomical and
phylogenetic claims, and counterclaims, will continue to
be insecure without further information about the topol-
ogy, composition and taphonomic history of anatomical
features, combined with explicit articulation of the meth-
odology used to reach anatomical interpretations.
MATERIALS AND METHODS
Topological reconstruction and comparative anatomy. In
order to address the problems of circularity and equivo-
cal interpretations, a stepwise methodology separating
topological considerations from anatomical interpreta-
tion was applied as advocated by Donoghue and Purnell
(2009). First, the features of the fossils were identified
and described (i.e. number and shape of distinct body
parts, and the topological relationship between those
body parts). By comparing specimens preserved in dif-
ferent orientations, three-dimensional reconstruction was
possible (cf. Briggs and Williams 1981; Purnell and
Donoghue 1999). Throughout this process of interpreta-
tion and reconstruction, no assumptions were made
about the affinities of the organism or the homology of
its features.
Following topological description, an explicitly justified
interpretative model was selected upon which to base
anatomical hypotheses. Putative homologies were identi-
fied through the consideration of topological relationships
between body parts (Rieppel and Kearney 2002) and were
informed by evolutionary and potential taphonomic
transformational sequences (i.e. assessing whether the
appearance of a feature, or its absence, represents the ori-
ginal anatomical condition, or the results of post-mortem
processes of decay and preservation). The intrinsic
properties and composition of the body parts provided
additional constraints on interpretations.
Following translation of topological structures into
anatomical interpretations, taxonomic assessments and
phylogenetic analyses were employed to investigate the
placement of the organism in an evolutionary context.
Elemental analysis. Topological data were complemented
by determination of the chemistry and preservation of
particular features to provide evidence of original histol-
ogy and ⁄ or composition (e.g. Butterfield 2002; Gabbott
et al. 2004). Elemental mapping of some features of Ja-
moytius was performed on a LEO 435 VP Scanning Elec-
tron Microscope with an Oxford Instruments ISIS 300
EDX spectrometer operating in variable pressure mode at
10 Pascals with an accelerating voltage of 10 kV and a
beam current of 500 picoamps for 1000 frames (approxi-
mately 14 h of run time). The specimens analysed in this
study have not been subjected to hydrofluoric acid treat-
ment (Ritchie 1963, 1968); smaller specimens were
selected because of SEM chamber size limits.
Phylogenetic analysis. Data matrices were constructed in
MacClade 4.06 (Maddison and Maddison 2003). Heuristic
searches were performed using PAUP 4.0 (Swofford 2002)
with 1000 random sequence addition replicates and TBR
(Tree bisection and reconnection) branch swapping.
Where appropriate, characters were reweighted according
to their rescaled consistency indices. To investigate the
alternative topologies that satisfy phylogenetic relation-
ships proposed on the basis of molecular data, heuristic
searches were conducted using backbone constraint trees
constructed in MacClade 4.06 (see below).
Institutional abbreviations and publicly held material. NHM, Nat-
ural History Museum, London, P11284 (holotype), P11285,
P47784-7; AMS, Australian Museum, Sydney, F64401, F102841-
6; NMS National Museum of Scotland, Edinburgh, 1959.1,
1966.3.1-3; BGS, British Geological Survey, Keyworth, 11882-3;
Hunterian Museum, Glasgow, V.7792, V.8036V.8141, V.8148,
GLHAM101382; UOE, University of Edinburgh, FR1628,
FR1476, 20129-32, 20145, 20159-62.
BODY PARTS, TOPOLOGICALANALYSIS AND RECONSTRUCTION
Body shape
In general aspect, Jamoytius has an elongate lozenge-
shaped body exhibiting a size range of 140–180 by 30–
EXPLANATION OF PLATE 1
Jamoytius kerwoodi White holotype (NHM P11284a) immersed in 90 per cent ethanol with incident polarized light and filter,
illustrating the conflicting interpretations of White (1946), in bold, and Ritchie (1960, 1963, 1968, 1984) in plain text. Scale bar
represents 10 mm.
1394 P A L A E O N T O L O G Y , V O L U M E 5 3
SANSOM et al., Jamoytius kerwoodi
eyemouth
intermuscle spaces(myocommata)interscale spaces
displaced skinlateral fin fold
basal supportsof anal finnot interpreted
lateral fin fold ventral terminationof body scales
notochordone marginof intestine
rays ofdorsal fin
not interpreted
bifurcation ofnotochord
branchial basket
eyeeye
basal supportsof dorsal fin
not interpreted
intestineone marginof intestine
not interpretedeye
lateral fin foldbranchial basket
muscle blocks(with muscle fibres)unmineralized scales(with ornamentation)
PLATE 1
40 mm. The body is preserved with a varying degree of
curvature (range of 40–90 degrees through the long axis
of the body). One end of the body of Jamoytius shows a
greater degree of morphological differentiation, contain-
ing multiple different substructures. This is provisionally
taken to be the head (see later discussion), thus indicating
anterior. Through comparison of paired and symmetrical
body parts, specimens are identified as collapsed remains
of a bilaterally symmetrical organism preserved in differ-
ent orientations (some are dorso-ventrally collapsed, some
laterally and others intermediate).
There is not enough evidence to determine the original
shape of the posterior of the organism, as most slabs do
not possess the most posterior portion. When present,
the posterior is preserved much more faintly than the rest
of the body, sometimes too faintly to be discerned clearly.
NHM P47784a exhibits what could be interpreted as two
posterior lobes, but the region is poorly preserved and
heavily prepared, thus obscuring the original body out-
line.
Body parts
Anterior subcircles. Four subcircles occur in the anterior
region. From comparison of specimens preserved in dif-
ferent orientations, it is apparent that they are coincident
with the body margin. Two distinct types of subcircle
occur. Two have broad, dark margins and form a lateral,
symmetrical pair, close to either the dorsal or ventral
body margin (Text-fig. 1). Comparison of the shapes of
these structures in dorso-ventral and laterally collapsed
specimens indicates that their original shape was either a
laterally flattened spheroid, or an outwardly opening cup-
like structure roughly equating to half or two-thirds of a
sphere (e.g. Text-fig. 1A). The other two subcircles have
narrower margins and lie along the sagittal plane, one ter-
minal, one subterminal in position (Text-fig. 1C,D). Irre-
spective of the orientations of the body, these axially
located rings are preserved as approximately circular out-
lines. This indicates either that they were originally spher-
ical, or that they were discs with sufficient rigidity at the
time of body collapse to reorient into a bedding parallel
attitude.
Elemental mapping analysis performed on the anterior
region of NHM P47787a shows a clear correlation of car-
bon with one of the paired, broader margined, anterior
subcircles (Text-fig. 2B).
Serial subrectangles. Towards the anterior end, many
specimens preserve a pair of linear features composed of
serially repeated, contiguous, subrectangular shapes
A B
C
A B C D
D
TEXT -F IG . 1 . Anterior subcircles of
Jamoytius (A–D) with corresponding
graphic interpretations (below), where
darker grey represents laterally paired
subcircles with broader margins, lighter
grey represents subterminal ring and
medium grey represents terminal ring.
A, NHM P11284a. B, NMS 1966.3.2. C,
NHM P11285. D, NMS 1966.3.1. Scale
bars represent 5 mm.
1396 P A L A E O N T O L O G Y , V O L U M E 5 3
(Text-fig. 3A,B). The central areas of the subrectangles
have the same coloration as the body but their perime-
ters are darker. Like the anterior subcircles, the lines of
subrectangles are coincident with the body margin. The
subrectangles are arranged in a ladder-like line, which
lies at a shallow angle to the antero-posterior axis of
the body.
The precise number of subrectangles is difficult to
establish and possibly varies between individuals. Jamoy-
tius has been reported to have had as many as 17 subrec-
tangles in each series (Ritchie 1984) and as few as seven
(Forey and Gardiner 1981). In the case of the specimen
discussed by the latter, the body is incomplete and the
posterior portion of the region with subrectangles may be
missing. On all specimens for which the entire length of
the line of subrectangle rows is present, at least 10 can be
identified, and on several, at least 14 (e.g. Text-fig. 3).
W-shaped structures. Although they are not preserved in
all specimens, among the most conspicuous features of
Jamoytius are W-shaped, serially repeated structures.
Their disposition indicates that they were coincident with
the body outline, or at least very nearly so (Pl. 1; Text-
fig. 4A). Each W extends around the majority of the lat-
eral body margin, leaving a gap on one body surface,
either dorsal or ventral (Pl. 1; Text-fig. 4D; Ritchie 1968,
pl. 4, fig. 1). The series of Ws does not extend along the
entire antero-posterior axis of the body; they are absent
from the anterior (e.g. Pl. 1) and their posterior limit is
uncertain. The Ws consist of alternating narrow and
broad zones.
The narrow zones are prominent and generally show
relief. They normally have a central area (200–300 lm
wide), the same colour as the matrix of the siltstone in
which the fossils are preserved, with very dark borders (c.
50 lm wide) on either side (Text-fig. 4C). In some
instances, the narrow zones also exhibit a tuberculate tex-
ture (Text-fig. 4E; Ritchie 1968, pl.4, fig. 3), which is best
observed on fragmentary specimens.
The broad zones (1–3 mm wide) lack relief, but are
darker in colour than other parts of the body. Within the
broad zones are linear features, lighter in colour; some of
these lines form dendritic patterns (Text-fig. 4C).
Elemental mapping of the W-shaped stripes of speci-
men GLAM V8141c shows that the borders of the narrow
zones and the whole of the broad zones contain associa-
tions of Ca and P (Text-fig. 2A) but not the other ele-
ments making up the matrix. Carbon is also present, but
appears to have an inverse distribution to that of Ca and
P; the W-shaped features are, therefore, interpreted as
being composed of calcium phosphate, possibly with an
organic component.
Axial lines and rounded structures. In the holotype (NHM
P11284a, Pl. 1, Text-fig. 5) and FR 1601 (Ritchie 1968; pl.
4, fig. 2, pl. 6, fig. 1), a pair of parallel, axial lines are
noted in the middle of the trunk (each approximately
2 mm wide). Upon closer inspection, the lines are com-
posed of contiguous lozenge- or oval-shaped units, each
slightly longer than wide (Text-fig. 6B). In FR 1601, the
region between the axial lines preserves the W-shaped
structures from both the near and far side of the body,
whilst the region outside of the axial lines preserves the
Ws from only one side of the body (Text-fig. 6A). In
NHM P11284a, one of the axial lines bifurcates towards
the anterior. Towards the posterior of NHMP 11284a, the
lines become less clear. Aligned with the axial lines are a
parallel and paired posterior series of dark, rounded
structures, which exhibit positive relief (Text-fig. 6B).
A B
C D
E F
TEXT -F IG . 2 . SEM back scatter and elemental maps. A–D,
W-shaped structures of GLAM V8141c (A, back scatter; B,
Carbon; C, Calcium; D, Phosphorous). E–F, Anterior subcircle
NHM 47787A (E, back scatter; F, Carbon).
S A N S O M E T A L . : T A P H O N O M Y A N D A F F I N I T Y O F T H E S I L U R I A N V E R T E B R A T E J A M O Y T I U S 1397
Their periodicity is approximately the same as that of the
W-shaped structures. The alignment of the axial rounded
structures with the axial lines, combined with the loz-
enge-shaped units of the lines, suggests that the rounded
structures and axial lines may comprise the same struc-
ture.
Paired longitudinal ‘folds’. A pair of parallel linear features
resembling collapsed folds is observed on the same sur-
face of the body as the gap between the W-shaped struc-
tures (Pl. 1, Text-fig. 3). They originate posterior to the
serial subrectangles and continue posteriorly until they
become indistinct. They are generally straight and exhibit
wrinkling in some instances.
Reconstruction
The general two-dimensional form of fossils of Jamoytius
is presumed to be a result of the collapse of a soft body
A
C
D
BA
B
TEXT -F IG . 3 . Anterior subrectangles
with corresponding graphic
interpretations (A, B) and ventro-lateral
‘folds’ (C, D). A, NMS 1966.3.2. B, NMS
1965.59a. C, NMS 1966.3.2. D, NHM
P11285. Scale bars represent 5 mm.
1398 P A L A E O N T O L O G Y , V O L U M E 5 3
during decay rather than compaction, and body fossils in
different orientations can thus be equated to two-dimen-
sional views of a three-dimensional organism (Briggs and
Williams 1981). Most of the preserved features are evi-
dent in almost all the dorso-ventrally and laterally col-
lapsed specimens that retain the appropriate portion of
the body, and the three-dimensional architecture and
position of these features can therefore be confidently
modelled. Whilst the model is a simplification of some
features (e.g. anterior subcircles), it corresponds well with
all known specimens of Jamoytius, including those that
are obliquely collapsed (Text-fig. 7). The accuracy of the
model can be tested by its ability to predict the position
of features in any newly discovered specimens.
The antero-posterior axis and dorso-ventral axis are
identified on the basis of the anatomical differentiation in
the ‘head’ and symmetrical disposition of surface struc-
tures (e.g. paired subcircles and W-shaped structures),
respectively; distinguishing dorsal and ventral remains
problematic in the absence of a phylogenetic context. The
model indicates that the paired axial lines and allied
rounded structures are likely to be interior structures.
They are preserved in only two specimens (NHMP 11284,
oblique; FR1601, lateral) but it seems that both lines
occur in the sagittal plane. Towards the anterior, one of
the axial lines is very close to the surface with no gap in
Ws towards the anterior; towards the posterior, the axial
lines approach the midline.
ANATOMICAL INTERPRETATION ANDCHARACTER HOMOLOGY
Establishing a phylogenetic context
Historically, Jamoytius has been interpreted as a jawless
vertebrate, but it lacks any unequivocal vertebrate synapo-
morphies (e.g. sensory canals, brain, skull, muscular phar-
ynx, multi-chambered heart, liver, kidney etc.) or, for
that matter, any unequivocal chordate synapomorphies
(e.g. dorsal nerve chord, notochord, myomeres, endo-
style ⁄ thyroid, pharyngeal arches, postanal tail). Paired
sense organs can be reasonably interpreted as present in
Jamoytius, but these are not unique to chordates
(although within chordates, they are a vertebrate synapo-
morphy). Previous interpretations of Jamoytius as a verte-
brate or even chordate have, therefore, not been
adequately justified: anterior anatomical differentiation,
serial stripes, axial lines, and a fusiform and curved body
shape are not sufficient in themselves to support a chor-
date model. Such general conditions could be noted in a
broad range of metazoan taxa, for example, articulated
A
C D E
B
dendritic pattern broad zone narrow zone displacement gap fractures tubercles
TEXT -F IG . 4 . W-shaped serially repeating stripes on the trunk of Jamoytius. A, NHM P11284a illustrating coincidence of stripes
with the left body margin. B, GLAM 101283 ⁄ 1. C, GLAM V8141. D, NHM P47784a. E, NHM P11284a. Scale bars represent 5 mm
(A–D) or 1 mm (E).
S A N S O M E T A L . : T A P H O N O M Y A N D A F F I N I T Y O F T H E S I L U R I A N V E R T E B R A T E J A M O Y T I U S 1399
soft-bodied fossils such as the purported polychaete Piec-
konia (e.g. Fitzhugh et al. 1997, fig. 7A.18). We can, how-
ever, compare Jamoytius with fossil or extant taxa that
possess the particular topological features outlined above.
Following redescription of Euphanerops (Janvier and
Arsenault 2007), it is clear that it shares with Jamoytius
the following features: dark anterior subcircles (specifi-
cally, a lateral pair, one terminal ring and one subtermi-
nal ring) and ladder-like rows of contiguous serially
repeating anterior subrectangles (Text-fig. 8). Two other
structures are comparable between the two genera but are
not present in exactly the same condition: serially repeat-
ing antero-posterior stripes and paired ventro-lateral
‘folds’. Unlike Jamoytius, Euphanerops also preserves some
unequivocal chordate synapomorphies (postanal tail,
terminal subcircle
subterminal subcirclelateral subcircle (l)
serial sub-rectangles (l)
serial sub-rectangles (r)
lateral subcircle (r)
ventro-lateralfolds
W-shaped structures
axial line (l/d)
axial line (r/v)
linear roundedstructures
TEXT -F IG . 5 . Body parts and topological interpretation of the
holotype (NHM P11284a) of Jamoytius. Scale bar represents
10 mm.
ventralaxialline
dorsalaxialline
Ws on right(proximal)surface
Ws on left(distal) surface
dorsalaxialline
ventralaxialline
anteriorbifurcation
axial roundedstructures
Ws on dorso-lateralsurface
A A
B B
TEXT -F IG . 6 . Axial structures of Jamoytius. A, Trunk of FR
1601 with reconstruction illustrating paired axial lines. B, Trunk
of NHM P12284a with reconstruction illustrating paired axial
lines and axial rounded structures. A1 from Ritchie (1968, pl. 6,
fig. 1). Scale bars represent 5 mm.
1400 P A L A E O N T O L O G Y , V O L U M E 5 3
notochord) and vertebrate synapomorphies (mineralized
endoskeleton, anal fin with fin supports) (Janvier and
Arsenault 2007). Given the similarities between Jamoytius
and Euphanerops in body parts and their topological
relations, it is reasonable to infer that Jamoytius also
possessed a postanal tail and notochord, which are not
A B
C
TEXT -F IG . 7 . Different perspectives of the three-dimensional model compared with two-dimensional fossil specimens of Jamoytius
preserved in different orientations. A, Dorsal perspective with NHM P11284a. B, Ventral perspective with NMS 1966.3.2. C, Lateral
perspective (anterior-posterior axis slightly oblique) with NHM P11285. Scale bar represents 5 mm.
A
B
C
contiguous subrectangles
median sub-terminal subcircle
lateralsubcircles
TEXT -F IG . 8 . Anatomy of
Euphanerops and Jamoytius. A, MHNM
01-02 drawing illustrating some of the
features shared with Jamoytius. B,
Reconstruction of Euphanerops. C.
Reconstruction of Jamoytius in light of
new data. A from Janvier and Arsenault
(2007, fig. 3B2), B adapted from Janvier
and Arsenault (2002, fig. 1b). Scale bar
represents 10 mm (A).
S A N S O M E T A L . : T A P H O N O M Y A N D A F F I N I T Y O F T H E S I L U R I A N V E R T E B R A T E J A M O Y T I U S 1401
preserved in known specimens. We reject the alternative
hypothesis that the similarities between Jamoytius and
Euphanerops are mere coincidences, on grounds of parsi-
mony. A chordate context for the interpretation of the
body parts of Jamoytius is thus justified, and topology can
be translated into anatomy in the light of this compara-
tive model (Table 1).
Anterior subcircles and subrectangles
Anatomical interpretation. Given a chordate model, the
lateral paired anterior subcircles are best interpreted as
optic capsules. The positions of the optic capsules of
Jamoytius can be interpreted as dorso-lateral given that in
all chordates with eyes they are located on either lateral
or dorso-lateral surfaces (rare exceptions include the ven-
tro-lateral eyes of the bighead carp, Hypophthalmichthys
nobilis). The dorsal and ventral surfaces of Jamoytius are
thereby identified – the optic capsules are dorsal, and the
gap between the W-shaped structures is along the ventral
surface.
Of the median anterior subcircles, the large rounded
subterminal ring, which is now clarified as being located
on the ventral body surface, is best interpreted as the oral
opening. Its position and architecture are consistent with
its interpretation as an annular cartilage (e.g. Ritchie
1963, 1968), such as that found in the extant lampreys.
There are, however, no associated structures preserved
that might support that hypothesis (e.g. circum-oral teeth,
copular cartilages, oral papillae). The smaller, terminal
subcircle is comparable to the single nasal opening
observed in a number of jawless vertebrates, both extant
and extinct. Its terminal position indicates that it is unli-
kely to be a pineal organ.
The paired, serially repeating subrectangles compare
closely to the branchial openings observed in fossil jawless
vertebrates such as Euphanerops, osteostracans, and, to a
lesser extent, extant jawless vertebrates such as lampreys.
These animals possess a series of external branchial
openings, which originate in the head region and descend
ventrally towards the posterior. In Jamoytius, the subrec-
tangles have previously been interpreted as internal, akin
to the branchial basket of lampreys (Ritchie 1963, 1968;
Forey and Gardiner 1981; Janvier 1981) and as such
would be lateral to the gills. The coincidence of the struc-
tures with the body margin makes them more comparable
with cartilaginous trematic rings surrounding the external
branchial openings. The Jamoytius subrectangles are, how-
ever, contiguous and numerous, unlike the trematic rings
of lampreys.
The apparent variability in the number of paired bran-
chial openings (10 or more pairs) may be because of the
nature of preservation of these features but could also
reflect real differences in the number of branchial struc-
tures; intraspecific variation in the number of branchial
units occurs in some jawless vertebrates (i.e. hagfishes).
Taphonomy and composition. Since Ritchie’s (1963)
description of Jamoytius, the features of the anterior have
generally been accepted as components of a cartilaginous
endoskeleton in part because of their inferred decay resis-
tance and similarity to that of lampreys. Jawless vertebrate
cartilages are quite varied in composition, both within
and between clades (e.g. Wright et al. 1998; Zhang et al.
2006), and this variability affects their decay resistance
(Sansom et al. 2010b). There are no definite fossil prece-
dents for the preservation of cartilage as organic films
(Euphanerops remains equivocal (Janvier and Arsenault
2002, 2007), whilst interpretations for conodonts (e.g.
Aldridge and Theron 1993) have been made through
comparison with Jamoytius). This does not in itself rule
out Ritchie’s (1963, 1968) interpretation of a cartilaginous
endoskeleton in Jamoytius, however. Without analytical
determination of the biomolecular composition of these
features (which may be impossible in these fossils), their
interpretation must rely solely on comparative anatomy
and comparative taphonomy.
The uniform preservation of the anterior structures as
dark, flat films, coupled with their carbonaceous compo-
sition, is consistent with organic preservation. Further-
more, their wrinkles and folds indicate flexibility at the
time of collapse; the absence of evidence of brittle defor-
mation indicates they were not rigid. Ductile deformation
need not rule out their interpretation as cartilaginous
supports for body openings (e.g. annular and trematic
rings), because cartilage, including that of jawless
vertebrates, can exist in both rigid and flexible forms,
but it does prompt consideration of other potential
body margin-related biomolecules. For example, high
concentrations of melanin are found in association with
TABLE 1 . Topological features identified in Jamoytius and
their anatomical interpretations based upon a chordate compar-
ator.
Topological feature Anatomical interpretation
Anterior subcircles
(paired, lateral)
Optic capsules
Anterior subcircle
(terminal)
Single, terminal, nasal
opening
Anterior subcircle
(subterminal)
Round ventral mouth
Anterior subrectangles Multiple external branchial
openings
W-shaped structures Rigid (probably mineralized)
scales
Axial lines with subunits Axial skeleton
Ventro-lateral paired ‘folds’ Lateral fin folds?
1402 P A L A E O N T O L O G Y , V O L U M E 5 3
photosensory structures of lampreys (e.g. optic capsules,
pineal organ, lateral line system (Young 1981)) and other
body openings (e.g. branchial (Bagenal 1973)). The distri-
bution of melanin in lampreys is, therefore, consistent
with the interpretation of the anterior structures of
Jamoytius having originally had a high melanin content.
Currently, we are unable to determine whether the
anterior structures of Jamoytius were melanin or cartilage.
Regardless of whether they represent cartilaginous sup-
ports for body openings or skin pigment surrounding the
body openings, the anterior subcircles and subrectangles
are still best interpreted as a mouth, nasal opening, eyes
and external branchial openings.
W-shaped structures
Anatomical interpretations. A number of fossil jawless ver-
tebrates exhibit serial V-, W- or Z-shaped bands along
the body. These have been interpreted either as myomeres
(e.g. Haikouichthys, conodonts) or external dermoskeletal
scales (osteostracans, certain anaspids). Similarly, the W-
shaped structures of Jamoytius have been interpreted as
muscle blocks (Forey and Gardiner 1981; White 1946)
and as scales, either mineralized (Ritchie 1960) or unmin-
eralized, ‘horny’ and carbonized (Ritchie 1968, 1984).
Others regard any interpretations of these structures as
equivocal (Janvier 1981).
The Ws of Jamoytius appear to be single units, rather
than a series of subunits arranged in a W-shape, as is the
case in the scales of most vertebrates. Euphanerops also
shows long, undivided structures that run the height of
its flank, yet it is uncertain whether they are scales or
myomeres (Janvier and Arsenault 2007).
Taphonomy and composition. The narrow zones of the W-
shaped structures have undergone brittle deformation in
the form of fracturing and displacement in several speci-
mens (Text-fig. 4D), thus indicating that they were rigid
prior to collapse or compaction. Interpretation of the Ws
as scales is supported by their rigid nature, tubercular
ornamentation, preservation in relief and coincidence
with the body margin.
The W-shaped structures are phosphatic and rigid,
demonstrating a very different taphonomic history to the
anterior features (carbonaceous composition and ductile
deformation). It is important to consider whether the
phosphate of the Ws is primary or secondary. Within the
Jamoytius horizon, primary phosphate occurs, for exam-
ple, in the dermal denticles of the thelodont Loganellia
(Marss and Ritchie 1998) and secondary phosphate is
known in the form of fibrous mineralized muscle in the
arthropod Ainiktozoon (Van Der Brugghen et al. 1997).
The texture and colour of the phosphate of Jamoytius
does not directly compare to either of these phosphates,
so its nature remains unclear. The fracturing of the scales
does, however, support the hypothesis that the scales were
biomineralized in vivo, prior to their deformation.
No evidence is found for the muscle fibres identified
by White (1946). It seems likely that this was a misinter-
pretation of the dendritic pattern of the broad zone
(Ritchie 1968), which cannot be reconciled with any fea-
ture of a myomere (Text-fig. 4C). The dendritic pattern
occurs on specimens that have not been treated with hy-
droflouric acid (contra Forey and Gardiner 1981), but not
on any specimen that exhibits tubercles. The variation in
the appearance of scales among specimens may, therefore,
relate to the level of the splitting of the scale passing
through different hard tissues. This is observed in some
osteostracans (R. Sansom, pers. obs.), in which the mid-
dle layer of the dermoskeleton can be exposed revealing a
dendritic pattern of ‘intra-areal’ canals (e.g. Denison
1947; Janvier 1996a; Sansom 2008). The relative thinness
of the scales of Jamoytius, however, is difficult to reconcile
with the hypothesis that the broad zone dendritic pattern
represents a canal system. Alternatively, the pattern could
be an artefact caused by taphonomic processes such as
fracture because of post-mortem shrinkage of the broad
zones.
Axial lines with subunits
Anatomical interpretation. The linear, dorsal and ventral
axial lines and their continuation posteriorly as lines of
rounded axial features should be assessed through com-
parison with antero-posterior axial structures known in
vertebrates, i.e. notochord, dorsal nerve cord, gut and
vertebral elements. The contiguous lozenge ⁄ oval subunits
are not consistent with previous interpretations as a noto-
chord and a gut (White 1946), or margins of a gut
(Ritchie 1968). A further inconsistency is the anterior
bifurcation of one of the lines (dorsal) that, contra to
Ritchie (1968), is not part of the branchial basket (Text-
fig. 5).
The internal pattern of subunits within the lines is
more in keeping with interpretation as an axial skeleton.
The contiguous nature of the subunits towards the ante-
rior and increasing separation towards the posterior is
comparable to the condition of the arcualia in lampreys
(Marinelli and Strenger 1954). The rounded or lozenge
shape of the elements in Jamoytius does not, however,
match the irregularly shaped cartilaginous arcualia of
either lampreys (a single series dorsal to notochord) or
Euphanerops (dorsal and ventral series, Janvier and Ar-
senault 2007, fig. 16). Rather, they are more comparable
to the ‘haemal series’ of Euphanerops, more specifically
the lozenge-shaped subunits of the posterior haemal
S A N S O M E T A L . : T A P H O N O M Y A N D A F F I N I T Y O F T H E S I L U R I A N V E R T E B R A T E J A M O Y T I U S 1403
series. Although the notochord of modern jawless verte-
brates (hagfish and lampreys) is wide, the gap between
the dorsal and ventral axial lines of Jamoytius is propor-
tionally far larger, potentially making interpretation as
dorsal and ventral arcualia problematic. Furthermore, the
anterior bifurcation of the dorsal line in NHM P12284 is
inconsistent with interpretation as arcualia.
Whilst it is not possible to determine precise homology
of the dorsal and ventral axial lines because of their unu-
sual shape and position, it is likely that they are formed
by subunits of some form of axial skeleton, either arcualia
in a previously unobserved condition or a ‘haemal series’
comparable to that of Euphanerops (Janvier and Arsenault
2007).
Taphonomy and composition. If interpretation of the dor-
sal and ventral axial lines of Jamoytius as axial skeleton is
accepted, then comparison with extant jawless vertebrates
indicates the subunits were likely composed of cartilage.
The seemingly different nature of the preservation of the
axial lines from that of the anterior subcircles ⁄ subrectan-
gles does not mean that they cannot both be composed
of cartilage: lamprey branchial cartilages have a different
composition from arcualial and neurocranial cartilages
(Fernandes and Eyre 1999; Robson et al. 1997).
Notochords interpreted in fossil jawless vertebrates
such as Gilpichthys (Bardack and Richardson 1977) and
conodonts (Aldridge et al. 1993) have a banded appear-
ance, seemingly because of overprinting of the myomeres.
In Jamoytius, evidence of myomeres is not preserved.
Whilst the W-shaped scales have a periodicity similar to
that of the axial line subunits, there are instances of the
narrow zones overlying the subunits. The subunits of the
axial lines of Jamoytius are, therefore, unlikely to be a
taphonomic artefact because of scales or myomere over-
printing. In FR 1601, the region between the dorsal and
ventral axial lines is the only region of the body to pre-
serve the W-shaped scales from both lateral sides of the
body (Text-fig. 6A). A similar pattern is observed in late-
stage decay of larval lampreys (e.g. Sansom et al. 2010a),
which, when viewed laterally, reveal myomeres from both
lateral sides of the body within the region occupied
by the wide notochord, but not the dorsal and ventral
sections of the body (R. Sansom, pers. obs.).
Paired longitudinal ‘folds’
Anatomical interpretation. The paired parallel ‘folds’ on
the ventro-lateral body margins (Text-fig. 3) have been
the source of conflicting interpretations. Some authors
regard them as ‘lateral fin folds’ (Janvier 1981; Ritchie
1968; White 1946), whilst others find no evidence to sup-
port that view (Forey and Gardiner 1981; Westoll 1958).
Long, thin, paired appendages (lateral fin folds) do not
occur in any form in extant vertebrates (Bemis and
Grande 1999), so we have compared the structures in
Jamoytius to those known in fossil jawless vertebrates.
Pharyngolepis (an anaspid) and Euphanerops possess long,
thin, paired ‘appendages’, which extend along the ventro-
lateral surfaces from the head to the anal region (Janvier
and Arsenault 2007; Ritchie 1964). In both of these taxa,
there are mineralized components of the ventro-lateral
appendages. Jamoytius lacks such mineralized structures,
and the ventro-lateral ‘folds’ are probably simple folds of
the skin.
Taphonomy and composition. Interpretation of the dorso-
ventral ‘folds’ as folds of skin without any form of
mineralized supports raises the question whether the
paired features are true anatomical features or a tapho-
nomic artefact. If the majority of the body surface was
covered with rigid scales (W-shaped structures), the more
flexible ventral surface in the gap between scales would be
prone to deformation during collapse. This hypothetical
taphonomic scenario is perhaps supported by the wrin-
kling that occurs within the ventro-lateral folds of some
specimens, but their paired nature argues against it
(Text-fig. 3). It is therefore unclear whether these folds
are distinct anatomical structures or merely a conse-
quence of body collapse; homologizing them with the
‘appendages’ of Pharyngolepis and Euphanerops is thus
currently problematic.
PHYLOGENETIC ANALYSIS
The evidence presented above indicates that Jamoytius is a
jawless vertebrate of uncertain affinities, so we have used
the most recent and comprehensive analysis of early ver-
tebrate interrelationships (Gess et al. 2006) as a basis to
investigate its precise phylogenetic position. The matrix of
Gess et al. (2006) is based upon earlier matrices (Janvier
1996b; Donoghue et al. 2000; Donoghue and Smith
2001), updated and expanded to include subsequently dis-
covered soft-bodied vertebrates, namely Haikouichthys
and Myllokunmingia (Shu et al. 1999; Donoghue et al.
2003; Hou et al. 2002), Mesomyzon (Chang et al. 2006)
and Priscomyzon (Gess et al. 2006) as well as oral charac-
ters relating to cyclostome monophyly.
Coding. The coding used for Jamoytius by Donoghue
et al. (2000) and subsequently by Donoghue and Smith
(2001) and Gess et al. (2006) was based upon an earlier
unpublished version of the data presented here (Freed-
man 1999), not all of which has survived subsequent
scrutiny. Thus, the coding has been modified to reflect
the interpretations herein (e.g. Table 1; Appendix S1).
1404 P A L A E O N T O L O G Y , V O L U M E 5 3
The matrix has also been updated to include additional
data for Euphanerops (Janvier and Arsenault 2007), Hai-
kouichthys (Zhang and Hou 2004), Arandaspida (Sansom
et al. 2005) and Galeaspida (Wang et al. 2005). Euphaner-
ops is taken to include Legendrelepis and Endiolepis (Jan-
vier 1996c; Janvier and Arsenault 2007). Two taxa from
the Middle Devonian of Scotland that have been pro-
posed to have affinities with Jamoytius have been added
to the matrix: Cornovichthys (Newman and Trewin 2001)
and Achanarella (Newman 2002).
Gess et al. (2006) employed presence ⁄ absence coding.
Such coding methodology violates the requirements of
logical independence of characters and can lead to false
support for a cladogram (Strong and Lipscomb 1999;
Forey and Kitching 2000). For example, absence of a nas-
ohypophyseal opening is counted twice in two different
characters (15 ⁄ 16), as is absence of dentine (80 ⁄ 81). The
matrix presented here therefore utilizes contingent coding,
which necessitates the erection of additional characters
(Appendix S1: characters 100–109). Furthermore, revision
of the coding and coding strategy for the characters relat-
ing to the relationships of extant cyclostomes reveals that
many of the physiological and miscellaneous characters
are uninformative (P. C. J. Donoghue, unpublished data).
These characters are removed here, whilst some of the
neurological characters are revised (e.g. cerebellar primor-
dia, retina).
Results. Heuristic searches, including all taxa, found two
most parsimonious trees of branch length 185 (Text-
fig. 9B). Jamoytius is placed as sister taxon to Euphaner-
ops, united by a ventral mouth and annular cartilage
(both homoplastic characters). These taxa (which could
together be termed Jamoytiiformes Tarlo, 1967) are
resolved as stem-gnathostomes because of their trunk der-
mal skeleton, separate anal fin and paired fin folds.
Ostraco/JawedOstraco/Jawed
TunicataCephalochordataMyxinoideaMyxinikelaHaikouichthys
PetromyzontidaMesomyzonPriscomyzonMayomyzonEuconodonta
JamoytiusEuphaneropsAnaspidaLoganelliaTuriniaHeterostraciArandaspidaAstraspisGaleaspidaOsteostraciJawed Vertebrates
1
1
11
1
1
2
2
2
1
1
13
1
Achanarella
Cornovichthys
1
1
2
MyxinoideaPetromyzontida
AnaspidaJamoytius
Ostraco/Jawed
MyxinoideaPetromyzontida
AnaspidaJamoytius
Ostraco/Jawed
Euphanerops
MyxinoideaPetromyzontida
AnaspidaJamoytius
Ostraco/Jawed
Euphanerops
EuconodontaPteraspidimorphi
MyxinoideaPetromyzontida
AnaspidaJamoytiusEuphaneropsEuconodonta
MyxinoideaPetromyzontida
AnaspidaJamoytiusEuphanerops
EuconodontaPteraspidimorphi
MyxinoideaPetromyzontida
AnaspidaJamoytius
Ostraco/Jawed
EuphaneropsEuconodonta
Forey 1995 Janvier 1996b Donoghue et al. 2000 Donoghue et al. 2001 Shu et al. 2003 Gess et al. 2006A
B C TunicataCephalochordata
MyxinoideaMyxinikela
Haikouichthys
PetromyzontidaMesomyzonPriscomyzonMayomyzonEuconodonta
JamoytiusEuphaneropsAnaspidaLoganelliaTuriniaHeterostraciArandaspidaAstraspisGaleaspidaOsteostraciJawed Vertebrates
1
1
1
1
1
1
1
1
2
1
1
12
1
Achanarella
Cornovichthys
1
1
2 1
1
TEXT -F IG . 9 . Phylogenetic relationships of Jamoytius. A, simplified versions of previous cladistic analyses of Forey (1995), Janvier
(1996b), Donoghue et al. (2000), Donoghue and Smith (2001), Shu et al. (2003), and Gess et al. (2006) where Ostraco ⁄ Jawed
represents other ostracoderms and jawed vertebrates. B, single most parsimonious tree from the unconstrained phylogenetic analysis
with decay support indices. C, Strict consensus of trees resulting from analysis constrained for cyclostome monophyly with decay
indices.
S A N S O M E T A L . : T A P H O N O M Y A N D A F F I N I T Y O F T H E S I L U R I A N V E R T E B R A T E J A M O Y T I U S 1405
Whether Jamoytius preserves these latter two characters is
uncertain, but the same topology results when the paired
appendages of Jamoytius are coded as unknown. Jamoytii-
formes are placed closer to the root of total-group gnat-
hostomes than Anaspida because of an absence of dermal
head covering, dentine and lamellar aspidin. Close rela-
tionships between Jamoytius and Euphanerops have been
reconstructed in previous phylogenetic studies (Janvier
1996a–c; Donoghue et al. 2000; Donoghue and Smith
2001; Shu et al. 2003), but always as part of a clade with
Anaspida (Text-fig. 9A).
Of the other taxa proposed to have jamoytiiform affini-
ties, Cornovichthys is placed as a stem-vertebrate (in the
sense that Petromyzontida and Gnathostomata comprise
Vertebrata, whilst vertebrates and Myxinoidea constitute
Craniata (Janvier (1981)), whilst Achanarella is placed
as a stem-gnathostome in a more basal position than
(Jamoytius + Euphanerops). Neither taxon is, therefore,
resolved as part of a monophyletic Jamoytiiformes.
The revisions incorporated in the data matrix used here
also led to other changes in relationships among jawless
vertebrates. Changing the coding strategy for dentine and
odontodes has led to the thelodonts (represented here by
Loganellia and Turinia) being identified as closer to the
root of total-group gnathostomes than the pteraspidimor-
phi (represented here by Heterostraci, Arandaspida and
Astraspis) on the gnathostome stem lineage. Furthermore,
eucondonts are placed as sister taxon to fossil and extant
lampreys, making conodonts stem-petromyzontids. Lam-
preys and euconodonts are united by possession of trans-
versely biting teeth.
Morphological evidence, both neontological and palae-
ontological, consistently finds lampreys (Petromyzontida)
as more closely related to jawed vertebrates than hagfishes
(Myxinoidea), as is the case here (Løvtrup, 1977; Janvier,
1981; Forey, 1984; Khonsari et al. 2009). Molecular inves-
tigations, however, identify the lampreys as more closely
related to the hagfishes and thus support cyclostome
monophyly (e.g. Delarbre et al. 2002; Delsuc et al. 2006).
Phylogenetic analysis of our data matrix constrained for
cyclostome monophyly identifies a less parsimonious
solution (branch length 191) with a topology very similar
to that of the unconstrained analysis, differing only in
placement of the myxinoids and resolution amongst petr-
omyzontids (Text-fig. 9C). The Jamoytiiformes are still
recovered as stem-gnathostomes, whilst the euconodonts
are recovered as stem-cyclostomes.
EVOLUTIONARY IMPLICATIONS
Jamoytius is commonly considered to represent a primi-
tive member of a fossil or extant vertebrate clade, either a
primitive anaspid (e.g. Ritchie 1963) or an ancestral lam-
prey (e.g. Mallat 1984). Despite the fact that a number of
its characters are plesiomorphic for chordates or for ver-
tebrates, analysis here establishes sister taxon relationship
with Euphanerops. In our unconstrained analysis, Jamoyti-
iformes represent a new grade in the evolution of
stem-gnathostomes, after the evolution of a trunk dermal
skeleton but before the evolution of lamellar aspidin and
dermal head skeleton. Given the preservation of trunk
dermoskeleton of Jamoytius (W-shaped scales), it is rea-
sonable to assume that any head dermoskeleton would
also be preserved if it had existed. The coding for absence
of head dermoskeleton in Jamoytius, and subsequent
placement of Jamoytiiformes on the gnathostome stem,
therefore reflects phylogenetic absence rather than tapho-
nomic loss (see Donoghue and Purnell 2009 for a discus-
sion of alternative meanings of stem assignments). The
position of Cornovichthys as a stem-vertebrate is sup-
ported by only one character (anterior otic capsules), the
interpretation of which is equivocal in some taxa. The
stem placements of Cornovichthys and Achanarella are
likely to reflect taphonomic bias resulting from loss of
characters through post-mortem decay (Donoghue and
Purnell 2009; Sansom et al. 2010a, b). To further resolve
the relationships of the Jamoytiiformes and putatively
related taxa, reinterpretation of the relevant fossils is
required using the same principles as applied here for
Jamoytius.
The phylogenetic placement of conodonts as stem-lam-
preys or stem-cyclostomes is contrary to the hypotheses
from previous analyses in which conodonts are placed as
stem-gnathostomes (Donoghue et al. 2000; Donoghue
and Smith 2001). The new placement is not robust, but it
is the most parsimonious on the basis of the morphologi-
cal data analysed here. The instability of this result sug-
gests that addition of further soft-bodied taxa to
phylogenetic matrices of early vertebrates will potentially
affect hypotheses of euconodont affinity. Our phyloge-
netic analysis does not raise any doubts about the place-
ment of euconodonts within the vertebrates.
Other stem-gnathostome taxa such as Pteraspidomor-
phi, Galeaspida and Osteostraci are resolved to be closer
to the gnathostome crown than thelodonts. This proposal
differs from previous suggestions of thelodont sister
relationships with gnathostomes (Marss et al. 2007),
chondrichthyes (Turner 1991) or the group (Galeasp-
ida + Osteostraci + jawed vertebrates) (Donoghue and
Smith 2001). Given the limited number of thelodont taxa
included here, however, questions of thelodont affinity
remain open to further investigation.
Jamoytius has often been cited in defence of the lateral
fin fold theory (e.g. Jarvik 1980; Shubin et al. 1997), an
evolutionary developmental scenario in which the paired
appendages of jawed vertebrates derive from continuous
ventro-lateral fin folds of jawless vertebrates by the loss of
1406 P A L A E O N T O L O G Y , V O L U M E 5 3
the intermediate portion of the fin fold (Balfour 1876;
Thacher 1877; reviewed by Coates 1994; Bemis and
Grande 1999). The present study found no evidence for
any skeletal or muscular structures that would allow an
assessment of potential homologies with the paired fins of
jawed vertebrates. Furthermore, the antero-posterior skin
folds may represent taphonomic artefacts. Our results
indicate that the structures of anaspids, thelodonts and
potentially Jamoytius were acquired independently of
paired fins restricted to the pectoral region in Osteostraci
and Gnathostomata (Sansom 2009).
CONCLUSIONS
The study of the anatomy of problematic organisms can
be aided by the use of a methodology designed to sepa-
rate topological and morphological reconstruction from
anatomical interpretation and to gather as much informa-
tion as possible about the preserved features through
taphonomic analyses. The application to Jamoytius dem-
onstrates that it is a vertebrate, with preserved W-shaped
phosphatic scales, ten or more paired external branchial
openings, dorso-lateral optic capsules, a round ventral
mouth, a terminal nasal opening, and, potentially, dorsal
and ventral axial skeleton. Interpretations of paired fins
remain equivocal. Analyses of the phylogenetic affinity of
Jamoytius identify a sister taxon relationship with Eupha-
nerops. This clade, the Jamoytiiformes, is a primitive
group of stem-gnathostomes and does not form a clade
with the Anaspida.
Acknowledgements. This work was funded in part by a Natural
Environment Research Council grant (NE ⁄ E015336 ⁄ 1 to SEG
and MAP). KF was supported by an Overseas Research Student
award (ORS ⁄ 96014039) and by R. A. Freedman. Various people
are thanked for their assistance in enabling the study and loan of
material including Sir Frederick Stewart, Peder Aspen (University
of Edinburgh), Neil Clark (Hunterian Museum), Bobbie Paton,
Liz Hide and Mike Taylor (National Museum of Scotland), Sally
Young, Peter Forey and Martha Ritcher (Natural History
Museum, London), Robert Jones and Alex Ritchie (Australian
Museum) and Steve Tunnicliff (British Geological Survey). Tony
Milodowski and Paul Wetton (British Geological Survey) kindly
assisted with SEM analysis. We also appreciate the constructive
comments of three anonymous reviewers and Philip Donoghue,
which have allowed us to improve the manuscript.
Editor. Philip Donoghue
SUPPORTING INFORMATION
Additional Supporting Information may be found in the online
version of this article:
Data S1. Previous interpretations of the affinity of Jamoytius.
Appendix S1. Character list and matrix used in phylogenetic
analysis – an updated version of Gess et al. (2006) with neuro-
logical characters adapted according to P. Donoghue (unpub-
lished data).
Please note: Wiley-Blackwell are not responsible for the con-
tent or functionality of any supporting materials supplied by the
authors. Any queries (other than missing material) should be
directed to the corresponding author for the article.
REFERENCES
A L D R I D GE , R. J. and T H E R ON , J. N. 1993. Conodonts
with preserved soft tissue from a new Ordovician Konservat-
Lagerstatte. Journal of Micropalaeontology, 12, 113–117.
—— B R I G G S , D. E. G., S M I TH , M. P., C L A R KS O N , E. K.
and C L A R K, N. D. L. 1993. The anatomy of conodonts.
Philosophical Transactions of the Royal Society of London, Series
B, 340, 405–421.
B A G E N A L , T. B. 1973. Identification of British fishes. Houlton
Educational Publications, Amsterdam, 199 pp.
B A L F OU R , F. M. 1876. Monograph on the development of
elasmobranch fishes. Journal of Anatomy and Physiology, 10,
377–441.
B A R DA CK , D. and R I CH A R D S ON , E. R. 1977. New
agnathous fishes from the Pennsylvanian of Illinois. Fieldiana
Geology, 33, 489–510.
B E M I S , W. E. and G R A N DE , L. 1999. Development of the
median fins of the North American paddlefish (Polyodon
spathula), and a reevaluation of the lateral fin-fold hypothesis.
41–68. In A R R A T I A , G. and S C HU L T Z E , H. P. (eds).
Mesozoic fishes 2 – systematics and fossil record. Verlag
Dr. Friedrich Pfeil, Munchen, 604 pp.
B R I G G S , D. E. G. and C L A R KS O N , E. K. 1987. An enigmatic
chordate from the Lower Carboniferous Granton ‘shrimp-bed’
of the Edinburgh district, Scotland. Lethaia, 20, 107–115.
—— and W I L L I A M S , S. H. 1981. The restoration of flattened
fossils. Lethaia, 14, 157–164.
B UT T E R F I E L D , N. J. 2002. Leanchoilia guts and the interpre-
tation of three-dimensional structures in Burgess Shale-type
fossils. Paleobiology, 28, 155–171.
C H A N G , M.-M., Z H A N G , J.-Y. and M I A O , D. 2006. A lam-
prey from the Cretaceous Jehol biota of China. Nature, 441,
972–974.
C OA TE S , M. I. 1994. The origin of vertebrate limbs. Develop-
ment, 120 (Suppl. 1), 169–180.
D E L A R BR E , C., G A L L UT , C., BA R R I E L , V., J A N V I E R ,
P. and G A C HE LI N , G. 2002. Complete mitochondrial DNA
of the hagfish, Eptatretus burgeri: the comparative analysis of
mitochondrial DNA sequence strongly supports the cyclo-
stome monophyly. Molecular Phylogenetics and Evolution, 22,
184–192.
D E L S UC , F., B R I N K M A N N , H., CH O UR R O UT , D. and
PH I LI PP E , H. 2006. Tunicates and not cephalochordates
are the closest living relatives of vertebrates. Nature, 439,
965–968.
S A N S O M E T A L . : T A P H O N O M Y A N D A F F I N I T Y O F T H E S I L U R I A N V E R T E B R A T E J A M O Y T I U S 1407
D E N I S O N , R. H. 1947. The exoskeleton of Tremataspis. Amer-
ican Journal of Science, 245, 337–365.
D ON O G H UE , P. C. J. and PU R N E L L , M. A. 2009. Distin-
guishing heat from light in debate over controversial fossils.
BioEssays, 31, 178–189.
—— and S M I T H , M. P. 2001. The anatomy of Turinia pagei
(Powrie), and the phylogenetic status of the Thelodonti.
Transactions of the Royal Society of Edinburgh: Earth Sciences,
92, 15–37.
—— F OR E Y , P. L. and A L DR I DG E , R. J. 2000. Conodont
affinity and chordate phylogeny. Biological Reviews, 75, 191–
251.
—— S M I TH , M. P. and S A N S OM , I. J. 2003. The origin and
early evolution of chordates: molecular clocks and the fossil
record. 190–223. In DO N O GH U E , P. C. J. and S M I T H ,
M. P. (eds). Telling the evolutionary time: molecular clocks and
the fossil record. CRC Press, London, 288 pp.
F E R N A N D E S , R. J. and E Y R E , D. R. 1999. The elastin-like
protein matrix of lamprey branchial cartilage is cross-linked
by lysyl pyridinoline. Biochemical and Biophysical Research
Communications, 261, 635–640.
F I T Z H UG H , K., S R O K A , S. D., KR U T Y , S., H E N DE R -
S O N , M. B. and H A Y , A. A. 1997. Polychaete worms. 64–83.
In S HA B I CA , C. W. and HA Y , A. A. (eds). Richardson’s
guide to the fossil fauna of Mazon Creek. Northeastern Illinois
University, Chicago, 308 pp.
F O R E Y , P. L. 1984. Yet more reflections on agnathan-
gnathostome relationships. Journal of Vertebrate Paleontology,
4, 330–343.
—— 1995. Agnathans recent and fossil, and the origin of
jawed vertebrates. Reviews in Fish Biology and Fisheries, 5,
267–303.
—— and G A R D I N E R , B. G. 1981. J. A. Moy-Thomas and his
association with the British Museum (Natural History). Bulle-
tin of the British Museum (Natural History). Geology, 35, 131–
144.
—— and J A N V I E R , P. 1993. Agnathans and the origin of
jawed vertebrates. Nature, 361, 129–134.
—— and K I T CH I N G, I. J. 2000. Experiments in coding
multistate characters. 54–80. In S CO T L A N D, R. and
PE N N I N G T ON , R. T. (eds). Homology and systematics:
coding characters for phylogenetic analysis. Taylor and Francis ⁄Systematics Association, London, 217 pp.
F R E E D M A N , K. 1999. Aspects of the taphonomy of jawless ver-
tebrates. Unpublished PhD thesis, University of Leicester, 166
pp.
G A B BO T T , S. E., H O U, X.-G., N O R R Y , M. J. and S I V E T -
E R , D. J. 2004. Preservation of Early Cambrian animals of the
Chengjiang biota. Geology, 32, 901–904.
G E S S , R. W., CO A T E S , M. I. and R U BI DG E , B. S. 2006. A
lamprey from the Devonian period of South Africa. Nature,
443, 981–984.
H O U, X.-G., A L D R I D GE , R. J., S I V E T E R , D. J.,
S I V E T E R , D. J. and F E N G , X.-H. 2002. New evidence on
the anatomy and phylogeny of the earliest vertebrates. Proceed-
ings of the Royal Society of London. Series B, Biological Sciences,
269, 1865–1869.
J A N V I E R , P. 1981. The phylogeny of the Craniata, with partic-
ular reference to the significance of fossil ‘agnathans’. Journal
of Vertebrate Paleontology, 1, 121–159.
—— 1996a. Early vertebrates. Clarendon Press, Oxford, 393 pp.
—— 1996b. The dawn of the vertebrates: characters versus com-
mon ascent in the rise of current vertebrate phylogenies.
Palaeontology, 39, 259–287.
—— 1996c. The Miguasha ‘Anaspida’. 134–140. In S C H U -
L TZ E , H. P. (ed.). Devonian fishes and plants of Miguasha,
Quebec, Canada. Verlag Dr. Friedrich Pfeil, Munich, 374 pp.
—— and A R S E N A U L T , M. 2002. Calcification of early verte-
brate cartilage. Nature, 417, 609.
—— —— 2007. The anatomy of Euphanerops longaevus Wood-
ward, 1900, an anaspid-like jawless vertebrate from the Upper
Devonian of Miguasha, Quebec, Canada. Geodiversitas, 29,
143–216.
—— and B US C H , R. M. 1984. Jamoytius-like vertebrates from
the Lower Devonian Manlius Formation of New York State.
Journal of Vertebrate Paleontology, 4, 501–506.
J A N V I E R , P. and L UN D , R. 1983. Hardistiella montanensis
n.gen. et sp. (Petromyzontida) from the Lower Carboniferous
of Montana, with remarks on the affinities of lampreys.
Journal of Vertebrate Paleontology, 2, 407–413.
J A RV I K , E. 1980. Basic structure and evolution of vertebrates,
Vol. 2. Academic Press, London, 337 pp.
K HO N S A R I , R. H., L I , B., V E RN I E R , P., N OR T H C UT T ,
R. G. and J A N V I E R , P. 2009. Agnathan brain anatomy and
craniate phylogeny. Acta Zoologica, 90 (Suppl. 1), 52–68.
L Ø V T R UP , S. 1977. The phylogeny of Vertebrata. Wiley and
Sons, London, 330 pp.
M A DD I S O N , D. R. and M A DD I S ON , W. P. 2003. Mac-
Clade 4.06. Sinauer Associates, Sunderland, MA.
M A L LA T, J. 1984. Feeding ecology of the earliest vertebrates.
Zoological Journal of the Linnean Society, 82, 261–272.
M A R I N E L L I , W. and S T R E N G E R , A. 1954. Vergleichende
Anatomie und Morphologie der Wirberltiere, Heft I. Lampetra
fluviatilis. Franz Deuticke, Vienna, 78 pp.
M A R S S , T. and R I T C HI E , A. 1998. Articulated thelodonts
(Agnatha) of Scotland. Transactions of the Royal Society of
Edinburgh: Earth Sciences, 88, 143–195.
—— T U R N E R , S. and K A R A T A J �UT E - T A L I M A A , V. N.
2007. ‘‘Agnatha’’ II: Thelodonti. Handbook of Paleoichthyolo-
gy, Vol. 1B. Verlag Dr. Friedrich Pfeil, Munchen, 143 pp.
N E W M A N , M. J. 2002. A new naked jawless vertebrate from
the Middle Devonian of Scotland. Palaeontology, 45, 933–
941.
—— and T R E W I N , N. H. 2001. A new jawless vertebrate from
the Middle Devonian of Scotland. Palaeontology, 44, 43–51.
P UR N E L L , M. A. and D ON O G HU E , P. C. J. 1999. Flat-
tened fossils, physical modelling and the restoration of col-
lapsed skeletons. 91–99. In S A V A ZZ I , E. (ed.). Functional
morphology of the invertebrate skeleton. John Wiley & Sons,
Chichester, 706 pp.
R I E P P E L , O. and K E A R N E Y , M. 2002. Similarity. Biological
Journal of the Linnean Society, 75, 59–82.
R I T C H I E , A. 1960. A new interpretation of Jamoytius kerwoodi
White. Nature, 188, 647–649.
1408 P A L A E O N T O L O G Y , V O L U M E 5 3
—— 1963. Palaeontological studies on Scottish Silurian fish beds.
Unpublished thesis, University of Edinburgh.
—— 1964. New light on the morphology of the Norwegian
Anaspida. Skrifter Utgitt Av Det Norske Videnskaps-Akademi I
Oslo 1. Matematisk-Naturvidenskapelige Klasse, 14, 5–35.
—— 1968. New evidence on Jamoytius kerwoodi White, an
important ostracoderm from the Silurian of Lanarkshire,
Scotland. Palaeontology, 11, 21–39.
—— 1984. Conflicting interpretations of the Silurian agnathan,
Jamoytius. Scottish Journal of Geology, 20, 249–256.
—— 1985. Ainiktozoon loganese Scourfield, a protochordate from
the Silurian of Scotland. Alcheringa, 9, 115–142.
R O B S ON , P., W R I G H T, G. M., Y OU S O N , J. H. and
KE E L E Y , F. W. 1997. A family of non-collagen-based carti-
lages in the skeleton of the sea lamprey, Petromyzon marinus.
Comparative Biochemistry and Physiology Part B: Biochemistry
and Molecular Biology, 118, 71–78.
S A N S OM , I. J., D ON O G H UE , P. C. J. and A LB A N E S I , G.
2005. Histology and affinity of the earliest armoured verte-
brate. Biology Letters, 1, 446–449.
S A N S OM , R. S. 2008. The origin and early evolution of the
Osteostraci (Vertebrata): a phylogeny for the Thyestiida. Jour-
nal of Systematic Palaeontology, 6, 317–332.
—— 2009. Phylogeny, classification and character polarity of the
Osteostraci (Vertebrata). Journal of Systematic Palaeontology, 7,
95–115.
—— G A B BO T T , S. E. and P UR N E L L , M. A. 2010a. Non-
random decay of characters causes bias in fossil interpretation.
Nature, 463, 797–800.
—— —— —— 2010b. Decay of vertebrate characters in hagfish
and lamprey (Cyclostomata) and the implications for the
vertebrate fossil record. Proceedings of the Royal Society B-Bio-
logical Sciences, doi: 10.1098/rspb.2010.1641.
S H U , D.-G., CO N W A Y - M O R R I S , S., H A N , J., ZH A N G ,
Z.-F., Y A S UI , K., J A N V I E R , P., CH E N , L., Z H A N G , X.-
L., L I U , J.-N. and L I U , H.-Q. 2003. Head and backbone of
the Early Cambrian vertebrate Haikouichthys. Nature, 421,
536–539.
—— LU O , H.-L., C ON W A Y - M O R R I S , S., ZH A N G , X.-L.,
H U, S.-X., C H E N , L., H A N , J., Z H U, M. and CH E N ,
L.-Z. 1999. Lower Cambrian vertebrates from South China.
Nature, 402, 42–46.
S H U BI N , N., TA BI N , C. and C A R R O L L, S. 1997. Fossils,
genes and the evolution of animal limbs. Nature, 388, 639–
648.
S T R O N G , E. and L I P S C OM B , D. 1999. Character coding
and inapplicable data. Cladistics, 15, 363–371.
S W OF F OR D , D. L. 2002. PAUP*. Phylogenetic Analysis Using
Parsimony (*and other methods) Version 4. Sinuaer Associ-
ates, Sunderland, MA.
T A R L O , L. B. H. 1967. Agnatha. 629–636. In H A R L A N D ,
W. B. (ed.). The fossil record. London Geological Society,
London, 827 pp.
T H A CH E R , J. K. 1877. Median and paired fins, a contribution
to the history of vertebrate limbs. Transactions of the Connecti-
cut Academy of Arts and Science, 3, 281–310.
T U R N E R , S. 1991. Monophyly and interrelationships of the
Thelodonti. 87–119. In CH A N G , M.-M., LI U, Y.-H. and
Z HA N G , G.-R. (eds). Early vertebrates and related problems
of evolutionary biology. Science Press, Beijing, 514 pp.
V A N DE R B R U GG H E N , W., S CH R A M , F. R. and
M A R T I L L , D. M. 1997. The fossil Ainiktozoon is an arthro-
pod. Nature, 385, 589–590.
W A N G, N.-Z., DO N OG H UE , P. C. J., S M I T H, M. M. and
S A N S O M , I. J. 2005. Histology of the galeaspid dermoskel-
eton and endoskeleton, and the origin and early evolution of
the vertebrate cranial endoskeleton. Journal of Vertebrate
Paleontology, 25, 745–756.
W E S TO L L , T. S. 1958. The lateral fin-fold theory and the
pectoral fins of ostracoderms and early fishes. 180–211. In
W E S T O L L , T. S. (ed.) Studies on fossil vertebrates. The Ath-
lone Press, London, 263 pp.
W H I TE , E. I. 1946. Jamoytius kerwoodi, a new chordate
from the Silurian of Lanarkshire. Geological Magazine, 83, 89–
97.
W O O DW A R D , A. S. 1900. On a new ostracoderm (Euphaner-
ops longaevus) from the Upper Devonian of Scaumenac Bay,
Province of Quebec, Canada. Magazine of Natural Hitsory,
‘series 7, 5, 416–419.
W R I G HT , G. M., KE E LE Y , F. W. and DE M ON T , M. E.
1998. Hagfish cartilage. 160–170. In J ØR G E N S E N , J. M.,
L OM H OL T , J. P., W E BE R , R. E. and M A L T E , H.
(eds). The biology of hagfishes. Chapman and Hall, London,
578 pp.
Y O U N G , J. Z. 1981. The life of vertebrates. Clarendon Press,
Oxford, 645 pp.
Z H A N G , G.-J., M I Y A M OT O , M. M. and C OH N , M. J.
2006. Lamprey type II collagen and Sox9 reveal an ancient ori-
gin of the vertebrate collagenous skeleton. Proceedings of the
National Academy of Sciences, 103, 3180–3185.
Z H A N G , X.-G. and H OU , X.-G. 2004. Evidence for a single
median fin-fold and tail in the Lower Cambrian vertebrate,
Haikouichthys ercaicunensis. Journal of Evolutionary Biology, 17,
1162–1166.
S A N S O M E T A L . : T A P H O N O M Y A N D A F F I N I T Y O F T H E S I L U R I A N V E R T E B R A T E J A M O Y T I U S 1409
1
SUPPLEMENTARY INFORMATION This chronological list (compiled by K. Freedman, post 1997 updated by R. Sansom) includes the published diagnoses and amendments to diagnoses of Jamoytius, as well as other descriptive/interpretative comments and suggested affinities. When a particular author has referred to Jamoytius in a similar way in more than one work, these references are covered together at the earliest date. An attempt has been made to include all references that presented new information, interpretations, or suggestions of affinities of Jamoytius, including all cladistic analyses. Not all of the many references to Jamoytius in textbooks, however, are itemized below, although some examples from popular texts are given. Woodward (on specimen label) wrote ‘Allied to Lasanius’. White (1946, p. 96) gave the following diagnosis: ‘Fossil Agnatha without armour or endoskeletal calcification; notochord persistent; simple lateral fin folds present; median fin fold represented by elongated dorsal and anal fins. Eyes probably simple, very large and anteriorly placed. Muscle-segments simple and numerous with a single flexure and undivided by horizontal septum. Mouth ? terminal; intestine short and straight.…rounded, rather elongate bodied but somewhat flattened head region; anterior margin of head transverse and eyes marginal. Dorsal fin continuous over hinder two-thirds of body; anal fin one-quarter as long, remote.’ White (1946) considered Jamoytius as ‘a conservative element of the main stock from which the various groups of craniate chordates arose’ (p. 95). Gregory (1951, p. 105) stated, ‘[Jamoytius] had long horizontal lateral fin-folds, running from behind the very small head to the base of the tapering tail, which, however, was not turned downward as it was in Anaspids. There was a long spineless dorsal fin and a shorter anal fin. The skin was apparently very thin and without armour. The internal skeleton seems to have been cartilaginous.’ He followed White (1946) in favouring Jamoytius as an ancestor to amphioxus [=Branchiostoma] and added that amphioxus might be related to the anaspids. Wängsjö (1952, p. 566) questioned White’s (1946) interpretation of the notochord and suggested that Jamoytius could be a larval or naked thelodont. Robertson (1953, pp 730, 734) described Jamoytius as an unarmoured form with paired fin folds. Robertson (1953) also said that Wängsjö placed Jamoytius with the Euphaneropidae in the anaspids. Wängsjö (1952), however, had argued that the shape of the head, with anteriorly placed eyes, excluded Jamoytius from the anaspids. White (1958) and Ritchie (1968) later quoted Robertson (1953) as assigning Jamoytius to euphaneropsid anaspids. Berg (1955, p. 25) classified Jamoytius as an anaspid, sharing an order with Endeiolepis aneri Stensiö, 1939 (but not Euphanerops). Berrill (1955, pp 200-215) accepted White’s (1946) interpretation and agreed in regarding Jamoytius as a primitive chordate, but proposed that the mouth was ventral and ‘considerably subterminal’ (p. 204) and that Jamoytius had amphioxus-like mouth structures and atrial cavity. Berrill (1955) also held that Jamoytius would have had lens-less eyes.
2
Denison (1956, p. 424) was ‘inclined to interpret this form quite differently [from White (1946)], considering it to be for its time an advanced, though not necessarily highly specialized, vertebrate. The absence of dermal armor, the fusiform body, the presence of long lateral and dorsal fin folds (if they really do exist), the highly developed metamerism, and the large eyes are all characters of a very active, fast-swimming vertebrate, functionally more progressive than most of its contemporaries.’ Lehman (1957, pp 175-176) reviewed White (1946) and Stensiö (1958), compared Jamoytius with Endeiolepis, and considered Jamoytius as a possible anaspid. Robertson (1957, p. 167) followed White’s (1946) interpretation. Stensiö (1958, pp 238-240) regarded Jamoytius as an anaspid and provided the following amended diagnosis, here translated from the French: (External branchial openings unknown.) Scales of the dorsal crest absent and dorsal swimming fold is probably differentiated in a long, dorsal fin. Dorsolateral ranges of scales of the flank are very thin, especially distinctive by their internal ribs and their ornamentation, of an ordinary type, with an angular flexure. Lateral swimming fold is probably without exoskeleton. Anal fin probably fairly long. (Ranges of dorsoventral pre-branchial scales unknown.). Stensiö 1964 (pp 171-172) repeated this diagnosis. Smith (1957, p. 394), referring to Stensiö’s (1958) paper while in the press, concurred with his interpretation of scales and assignment to the anaspids. Smith (1957) added that the structures interpreted by White (1946) as ‘myocommata are the thickened, basal, dorso-ventral ridges of the scales’. Westoll (1958, pp 196-197) considered Jamoytius as a chordate without an exoskeleton and with little or no endoskeleton. He discussed White’s (1946) interpretation and recommended that the features identified by White as lateral fin folds could be better interpreted as dorsal and anal fins. White (1958, p. 229) noted that his interpretations and suggested affinity for Jamoytius had not been widely accepted by palaeontologists and discussed the work of Gregory (1951), Wängsjö (1952), and Robertson (1953). White (1958), however, still argued that Jamoytius ‘was the conservative derivative of the ancestor of them all.’ Ritchie (1960, p. 649), based on newly discovered material and reexamination of the type specimens, judged Jamoytius to have ‘…exoskeletal ossification (scales); lateral eyes; terminal, subcircular mouth; branchial apparatus not unlike that in the Anaspida and the living cyclostomes; fin rays in the lateral fin-folds; a hypoceral (?) tail.’ Jamoytius ‘shows a close relationship with the Anaspida…but of all the known fossil Agnatha, Jamoytius would seem to be the closest to the living Cyclostomata.’ Tarlo (1960, p. 117) followed Stensiö’s (1958) attribution of Jamoytius to the anaspids and his consideration that White’s (1946) ‘muscles' were scales with anaspid-like ornamentation. Tarlo (1960) added that the structures described by White (1946) as dorsal fin rays were dorsal ridge scales, but he accepted White’s (1946) interpretation of the notochord and intestine.
3
Swinton (1961, p. 36) figured a reconstruction of Jaymoytius [sic], based on White (1946), as a possible ‘relation of the very early vertebrates’. Ritchie (1963, pp 50-65, figs 10.2-10.3, and pls 15-22) described an annular cartilage in a near terminal position, cartilaginous eye capsules, and a cartilaginous, horizontally-aligned, ventrolateral branchial basket lying just behind the eyes with up to 15 gill pouches. He identified a tentative nasal structure, scales with ‘posthumous cracking’ (p. 59) composed of a horny epidermal material and tubercles in the interscale spaces, a long and straight intestine, possible lateral fin folds, a possible dorsal fin, and a probably hypoceral tail. Jamoytius ‘resembles closely the known anaspids and the living cyclostomes’ (p. 64), so Ritchie (1963) regarded Jamoytius as an unusual and primitive member of the anaspids. Dechaseaux (1963, pp 325, 328-330) concluded that Jamoytius could definitely be attributed to the agnathan vertebrates and could not be an ancestor to amphioxus. Deschaseaux (1963) subscribed to Ritchie’s (1960) view that, of all known fossil aganthans, Jamoytius most closely compared to the extant agnathans, but could still relate to the anaspids. Lehman (1964, pp 83-85) reviewed White (1946), Stensiö (1958), Berrill (1955), and Ritchie (1960) and accepted Ritchie’s (1960) interpretation and suggestion of affinity. Romer (1966, pp 17, 347R; 1968, p. 25) and Romer and Parson (1986, pp 43, 231) classified Jamoytius as an anaspid. Romer (1966) noted the presence of bony scales. Romer and Parson (1986) recognized separate gill openings, a lamprey-like branchial basket, and a single median nostril. Carter (1967, pp 26, 38-39) regarded Jamoytius as an anaspid or their close relative and remarked that it had an anaspid-like ventrolateral fin. Ritchie (1968) diagnosed Jamoytius as having an annular cartilage, sclerotic cartilages, cyclostome-like branchial basket, lateral trunk scales of unusual structure and composition, continuous dorsal fin and lateral fin folds, and probable hypocercal caudal fin. He deemed Jamoytius a ‘…cephalaspidomorph agnathan closely related to the anaspids and the living cyclostomes’ (p. 21) Bardack and Zangerl (1968, p. 1267) reported that ‘Jaymoytius [sic] has cyclostome characters such as a sub-terminal mouth, circular mouth and a branchial basket that begins behind the orbit; but Jaymoytius [sic] retains lateral fin folds, body scales, lacks a piston, and has 15 or more gill slits.’ They considered Jamoytius a possible ancestor to the lampreys. Bardack and Zangerl 1971 (p. 82) also noted the possible presence of an annular cartilage and established Jamoytius ‘…closer to cyclostome ancestry than the more usual Devonian anaspids…’ Colbert (1969, p. 17) said that Jamoytius ‘appears to be a very primitive jawless vertebrate, perhaps, occupying a position close to the ancestry of the lamprey and its relatives. The few, rather enigmatic fossil remains of Jamoytius show that this animal was small, elongated and tubular shaped. It had a terminal suctorial mouth, and, on each side of the head region, there was a row of circular gill openings, behind the eye.
4
There was a propulsive tail fin with a long lower lobe and a shorter deeper lobe and possibly there were lateral fin folds and a long dorsal fin, for maintaining balance. Although Colbert (p. 25) discussed White’s (1946) suggested affinity, he concurred with Ritchie by placing Jamoytius with the anaspids (p. 18). Wickstead (1969) compared the branchial structures in Jamoytius with those of larval amphioxus and the tail of Jamoytius with that of Asymmetron. He thought Jamoytius might represent a larval or metamorphosing acraniate. Olson (1971, pp 6-7) described Jamoytius as ‘a soft-bodied, fish-like creature’. He noted that White (1946) had regarded it as a vertebrate ancestor, but that most workers agreed to associate it with the anaspids or thelodonts. Halstead and Turner (1973, p. 70) cited the presence of a cartilaginous branchial basket and a round mouth and classified Jamoytius as an anaspid. Norman and Greenwood (1975, pp 30, 348) thought that Jamoytius, probably an aberrant anaspid, provided the only direct fossil evidence for a continuous ventrolateral fin fold. Nelson (1976, pp 22-23) listed Jamoytius as a ‘virtually naked’ anaspid. Janvier and Blieck (1979, p. 293) wrote that ‘the status of Jamoytius kerwoodi White is unclear, but it is likely that it is more closely related to the Petromyzontida than to the typical Anaspidida’. Lehman (1980, pp 229-230) continued to follow Ritchie (1960) and reported Jamoytius to have two symmetrical eyes, a round mouth, 7 branchial openings, and a hypocercal tail. He mentioned Stensiö’s (1964) association of Jamoytius with the anaspids and Ritchie’s hestitation in such an assignment of affinity. Forey and Gardiner (1981, pp 139-140) produced an amended diagnosis, recognizing ‘A naked cyclostome with a diphypcercal tail and branchial basket.’ They found no trace of scales or lateral fin folds, but reported an annular cartilage, ‘a branchial basket…with horizontal struts and a diphypcercal tail. The branchial basket has no more than seven openings…’ They concluded that ‘In our estimation the fossil looks very similar to the present-day lamprey…’ Janvier (1981, p. 139) followed Ritchie’s (1960, 1968) interpretation, but held some reservations about the scales and questioned the number of branchial openings and as to whether the annular cartilage ‘might also be a large olfactory organ…’(p. 139). Janvier (1981) considered Jamoytius as a sister-taxon to either the lampreys or the anaspids, but excluded it (along with Endeiolepis) from the anaspids due to the presence of a dorsal fin. Young (1981, p. 111) depicted Jamoytius with a notochord, a hypocercal tail, scales, lateral fin folds, an annular cartilage, and up to 15 branchial pouches with a branchial basket. He thought it probably an anaspid, but also proposed that it might be an ammocoete larva of an ostracoderm.
5
Hardisty (1982, p. 234) recorded that Jamoytius, a possible ancestor to the lampreys, appeared to have traces of an annular cartilage and a branchial basket. Halstead (1982, pp 170-171, 190) supposed that Jamoytius had scales, well developed paired fins, and a lamprey-like branchial basket, for which ‘… the arrangement of the gill openings was not compressed as in the anaspids but more akin to the situation of the lampreys.’ (p. 170). Halstead’s (1982) ‘cladogram’ showed Jamoytius as a sister-taxon to the anaspids and, together, as a sister-group to the lampreys, though he discussed other possible affinities. Janvier and Lund (1983, p. 412) thought Jamoytius had only seven branchial openings, a branchial basket, probably (horny?) scales, and possibly an anaspid-like paired fin. They also remarked, ‘Even more problematical is the head of Jamoytius, which shows a very short preorbital region, much too short to be considered similar to lampreys. The anterior black mass regarded by Ritchie (1960, 1968) and Forey and Gardiner (1981) as an annular cartilage might also represent a large, subterminal olfactory organ’ (p.412). Their cladistic analysis resolved Jamoytius as a sister-taxon to the lampreys, and together as a sister-group to the anaspids. Janvier and Busch (1984, pp 503, 505) reported that Jamoytius had horny (or poorly mineralized) scales, a possible annular cartilage, an elongate body shape, and a reduced anal fin. They linked Jamoytius with un-named ‘Jamoytius-like vertebrates’ on the basis of chevron-shaped unmineralized scales and discussed possible affinities with the lampreys. They concluded, however, that its only similarities to anaspids were craniate or vertebrate plesiomorphies. Mallatt (1984, p. 267) considered Jamoytius as an ancestral lamprey. Ritchie (1984), responding to Forey and Gardiner (1981), supported his previous interpretations and figured a new specimen with 15-17 ‘branchial arches (or apertures)’ (p. 254). Ritchie (1984) still thought of Jamoytius as ‘probably a close relative, if not an ancestor, of the petromyzontids’ (p. 255), but he was not certain whether the presence of a dorsal fin warranted removal of Jamoytius from the Anaspida sensu stricto, as suggested by Janvier (1981). Forey 1984 (p. 336) did not regard Jamoytius as an anaspid. Maisey (1986, p. 207) considered Jamoytius as an anaspid with a dorsal fin and listed (p. 210) possible characters linking the anaspids and lampreys, including the following from Jamoytius: branchial basket, continuous dorsal fin, reduced ossification, and annular cartilage? (noting, on p. 211, Janvier’s (1981) uncertainity over the interpretation of this feature). Briggs and Clarkson (1987, p. 110-112, 114-115) referred to the presence of V-shape scales in Jamoytius, and, on that basis, submitted that Jamoytius could form a clade with the Jamoytius-like vertebrates and Conopiscius. Janvier (1987, p. 850) noted the presence of elongated lateral fin folds.
6
Carroll (1988, pp 39, 40 596L) catalogued Jamoytius as a questionable anaspid with reduced armour and lamprey-like circular gill pocuhes and annular cartilage. Smith and Hall (1990, pp 302-03) reviewed the scales versus muscles debate. Arsenault and Janvier (1991, pp 28-32) placed Jamoytius as a sister-taxon to the lampreys, as they considered that it had the following characters: paired fins, ‘horny’ scales, sinuous branchial arches bearing spiny processes and associated trematic rings, an annular cartilage, approximately 30 branchial openings, elongation of the body, a dorsal fin, and a possibly reduced anal fin. Aldridge et al. (1993, pp 409-410), Aldridge and Theron 1993 (p. 116), and Aldridge and Donoghue 1998 (p. 19) accepted the presence of sclerotic eye cartilages in Jamoytius and compared them with structures in the conodont animals. Aldridge et al. (1993) also likened features of the conodont animals to the branchial structures of Jamoytius. Briggs and Kear (1993, p. 285) found no evidence for biomineralized scales. ‘The clear lines delineating the segments are most likely to be cartilaginous connective tissues which make up the myosepta (myocommata of White 1946), the dark surface in between representing the skin.’ Forey and Janvier (1993, p. 132) figured a cladogram with Jamoytius as a sister-taxon to the lampreys, and both together as a sister-group to the anaspids. Forey and Janvier (1994, pp 556, 560) provided a cladistic analysis indicating Jamoytius as a sister-taxon to the lampreys. Forey (1995, pp 276, 284, 286-287, 289-291) considered Jamoytius to have an annular cartilage, large eyes, a branchial basket with sloping gill pouches, a hypocercal tail, and paired ventrolateral fin folds. His cladistic analysis showed Jamoytius as a sister-taxon to the lampreys. Northcutt (1996, p. 239) wrote, ‘There is a general consensus that the lampreys share a common ancestor with a small extinct fish, Jamoytius…, but it is presently unclear whether this radiation is closely related to a second group of jawless fish, the anaspids…’ Janvier (1996a, pp 265, 270, 273-274, 278) produced a cladistic analysis that fixed Jamoytius as a sister-taxon to Euphanerops, and the two together as a sister-group to the anaspids (Text-fig. 5B). Janvier (1996b, pp 101-104, 238, 241) described Jamoytius as an eel-shaped form with no mineralised skeleton, which may have possessed elongate paired fins. Some specimens show tarry imprints of the eyes….possibly the olfactory organ or an annular cartilage…and a branchial ‘basket’ with about 20 branchial units or openings….’(p. 101). His cladistic analysis placed Jamoytius as a sister-taxon to the lampreys, Jamoytius and the lampreys together as a sister-group to Euphanerops, and all three together as a sister-group to the anaspids.
7
Pough et al. (1996, pp 184, 194) described lateral fin fold theory as being based upon the presence of the structures in anaspids like Jamoytius. Shubin et al. (1997, figure 1) illustrate Jamoytius as a continuous fin fold stage in the evolution of vertebrate limbs. Janvier (1998, p. 944) mentioned Jamoytius as a form without a mineralized skeleton, considered by some as an ancestor to the lampreys. Dineley (1999, p.42-44) discusses the various conflicting interpretations of Jamoytius anatomy and affinity. Shu et al. (1999) undertake a phylogenetic analysis of Haikouichthys and relatives, identifying Jamoytius as related to lampreys rather than anaspids and construct the primitive vertebrate condition as possesing lateral fin folds. Bemis and Grande (1999, p.61) highlight interpretations of Jamoytius as an example of circularity involved in discussion of the lateral fin fold hypothesis. Donoghue et al. (2000) compare the anatomy of conodonts with Jamoytius and undertook a parsimony analysis based upon this comparison. Jamoytius is resolved in a clade with anaspids but sister to Euphanerops. Donoghue and Smith (2001) expanded the phylogenetic analysis of Donoghue et al. (2000), and resolved Jamoytius as belonging to a clade containing Euphanerops and Anaspida. Shu et al. (2003) amend their phylogenetic analysis and resolve Jamoytius as sister to Anaspida. Gess et al. (2006) undertake a phylogenetic investigation based upon their description of Priscomyzon and resolve Jamoytius as a stem-gnathostome, not belonging to a clade including Euphanerops or Anaspida. Janvier and Arsenault (2007, p. 204-5) compared Jamoytius to Euphanerops considering the nature of the scales and median stains. They consider interpretations equivocal and as such reach few conclusions regarding Jamoytius’ anatomy or affinity. REFERENCES FOR SUPPLEMENTARY INFORMATION ALDRIDGE, R. J. and THERON, J. N. 1993. Conodonts with preserved soft tissue
from a new Upper Ordovician Konservat-Lagerstätte. Journal of Micropalaeontology, 12, 113-117.
—— SMITH, M. P., CLARKSON, E. N. K. and CLARK, N. D. L. 1993. The anatomy of conodonts.Philosophical Transactions of the Royal Society of London, Series B, 340, 405-421.
ARSENAULT, M. and JANVIER, P. 1991. The anaspid-like craniates of the Escuminac Formation (Upper Devonian) from Miguasha (Québec, Canada), with remarks on anaspid-petromyzontid relationships.19-40. In CHANG, M.
8
M., LIU, Y. H., and ZHANG, C. (eds). Early vertebrates and related problems of evolutionary biology, Sciences Press, Beijing, 514 pp.
BARDACK, D. and ZANGERL, R. 1968. First fossil lamprey: a record from the Pennsylvanian of Illinois. Science, 162, 1265-1267.
—— —— 1971. Lampreys in the fossil record. 67-84. In HARDISTY, M. W. and POTTER, I. C. (eds). The biology of lampreys. Volume 1. Academic Press, New York, 423 pp.
BEMIS, W. E. and GRANDE, L. 1999. Development of the median fins of the North American paddlefish (Polyodon spathula), and a reevaluation of the lateral fin-fold hypothesis. In G. Arratia and H. P. Schultze (eds). Mesozoic Fishes 2 - Systematics and Fossil Record. Verlag Dr. Friedrich Pfeil, München, 41-68 pp.
BERG, L. 1955. Classification of fishes, both recent and fossil. Trudy Zoologicheskogo Instituta Akademiya Nauk, SSSR, 20, 1-286. [In Russian].
BERRILL, N. J. 1955. The origin of vertebrates. Clarendon Press, Oxford, 257 pp. BRIGGS, D. E. G. and CLARKSON, E. N. K. 1987. An enigmatic chordate from the
Lower Carboniferous Granton Shrimp-Bed of the Edinburgh District, Scotland. Lethaia, 20, 107-115.
—— and KEAR, A. 1994a. Decay of Branchiostoma: implications for soft-tissue preservation in conodonts and other primitive chordates. Lethaia, 26, 275-287.
CARTER, G. S. 1967. Structure and habit in vertebrate evolution. Sidgwick and Jackson, London, 520 pp.
CARROLL, R. L. 1988. Vertebrate paleontology and evolution. W. H. Freeman and Company, NewYork, 698 pp.
COLBERT, E. H. 1969. Evolution of the vertebrates: a history of backboned animals through time. John Wiley and Sons, Inc., New York, 535 pp.
DENISON, R. H. 1956. A review of the habitat of the earliest Vertebrates. Fieldiana, Geology, 11, 357-457.
DECHASEAUX, C. 1963. Jamoytius, Amphioxus et le debut de l'histoire des Vertebres. Annales de Paleontologie, 49, 325-330.
DINELEY, D. L. 1999. Silurian fossil fishes sites of Scotland. In D. L. Dineley and S. J. Metcalf (eds). Fossil Fishes of Great Britain. Joint Nature Conservation Committee, Peterborough, 31-62 pp.
DONOGHUE, P. C. J. and SMITH, P. 2001. The anatomy of Turinia pagei (Powrie), and the phylogenetic status of the Thelodonti. Transactions of the Royal Society of Edinburgh: Earth Sciences, 92, 15-37.
—— FOREY, P. L. and ALDRIDGE, R. J. 2000. Conodont affinity and chordate phylogeny. Biological Reviews, 75, 191-251.
GESS, R. W., COATES, M. I. and RUBIDGE, B. S. 2006. A lamprey from the Devonian period of South Africa. Nature, 443, 981-984.
FOREY, P. 1984. Yet more reflections on agnathan-gnathostome relationships. Journal of Vertebrate Paleontology, 4, 330-343.
—— 1995. Agnathans recent and fossil, and the origin of jawed vertebrates. Reviews of Fish Biology, 5, 267-303.
—— and JANVIER, P. 1993. Agnathans and the origin of jawed vertebrates. Nature, 361, 129-134.
—— —— 1994. Evolution of the early vertebrates. American Scientist, 82, 554-565. —— and GARDINER, B. G. 1981. J. A. Moy-Thomas and his association with the
British Museum (Natural History). Bulletin of the British Museum (Natural History), Geology Series, 35, 131-144.
9
GREGORY, W. K. 1951. Evolution emerging. The MacMillan Company, New York, 736 pp.
—— and TURNER, S. 1973. Silurian and Devonian ostracoderms. 67-79. In HALLAM, A. (ed.). Atlas of palaeobiogeography, Elsevier Scientific Publishing, Amsterdam, 531 pp.
HARDISTY, M. W. 1982. Lampreys and hagfishes: analysis of cyclostome relationships. 165-259. In HARDISTY, M. W. and POTTER, I. C. (eds). The biology of lampreys. Volume 4B. Academic Press, London, 466 pp.
JANVIER, P. 1981. The phylogeny of the Craniata, with particular reference to the significance of fossil ‘agnathans’. Journal of Vertebrate Paleontology, 1, 121-159.
—— 1987. The paired fins of anaspids - one more hypothesis about their function. Journal of Paleontology, 61, 850-853.
—— 1996a. The dawn of the vertebrates: characters versus common ascent in the rise of current vertebrate phylogenies. Palaeontology, 39, 259-287.
—— 1996b. Early vertebrates. Oxford Monographs on Geology and Geophysics No. 33. Clarendon Press, Oxford, 393 pp.
—— 1998. Les vertébrés avant le Silurien. Geobios, 30, 931-950. —— and ARSENAULT, M. 2007. The anatomy of Euphanerops longaevus
Woodward, 1900, an anaspid-like jawless vertebrate from the Upper Devonian of Miguasha, Quebec, Canada. Geodiversitas, 29, 143-216.
—— and BLIECK, A. 1979. New data on the internal anatomy of the Heterostraci (Agnatha), with general remarks on the phylogeny of the Craniota. Zoologica Scripta, 8, 287-296.
—— and BUSCH, R. M. 1984. Jamoytius-like vertebrates from the Lower Devonian Manlius Formation of New York State. Journal of Vertebrate Paleontology, 4, 501-506.
—— and LUND, R. 1983. Hardistiella montanensis n.gen. et sp. (Petromyzontidda) from the Lower Carboniferous of Montana, with remarks on the affinities of the lampreys. Journal of Vertebrate Paleontology, 2, 407-413.
LEHMAN, J. P. 1957. Un probleme non resolu: L' origine des Vertebres. La Nature, (3265), 174-177.
—— 1964. L'origine des Vertebres. Traite de Paleontologie, 4, 78-91. —— 1980. Les progres recents de la paleontologie des Vertebres. Livre jubilaire du
cent cinquantenaire 1830-1980, Memoire Hors Serie- Societe Geologique de France, 10, 227-244.
MAISEY, J. G. 1986. Head and tails: a chordate phylogeny. Cladistics, 2, 201-256. MALLAT, J. 1984. Feeding ecology of the earliest vertebrates. Zoological Journal of
the Linnean Society, 82, 261-272. NORMAN, J. R. and GREENWOOD, P. H. 1975. A history of fishes. Ernest Benn,
Limited, London, 467 pp. NORTHCUTT, R. G. 1996. The agnathan ark - the origin of craniate brains. Brain
Behavior and Evolution, 48, 237-247. OLSON, E. 1971. Vertebrate paleozoology. Wiley-Interscience, New York, 839 pp. POUGH, F. H., HEISER, J. B. and MCFARLAND, W. N. 1996. Vertebrate life.
Prentice Hall International, New Jersey, 798pp. RITCHIE, A. 1960. A new interpretation of Jamoytius kerwoodi White. Nature, 188,
647-649. —— 1963. Palaeontological Studies on Scottish Silurian Fish Beds. Unpublished
Ph.D. Thesis,
10
University of Edinburgh, Edinburgh. —— 1968. New evidence on Jamoytius kerwoodi White, an important ostracoderm
from the Silurian of Lanarkshire, Scotland. Palaeontology, 11, 21-39. —— 1984. Conflicting interpretations of the Silurian agnathan, Jamoytius. Scottish
Journal of Geology, 20, 249-256. —— 1985. Ainiktozoon loganense Scourfield, a protochordate from the Silurian of
Scotland. Alcheringa, 9, 115-142. ROBERTSON, G. M. 1953. Some attempts at phylogeny of early vertebrates. Iowa
Academy of Science, 60, 725-737. ROBERTSON, J. D. 1957. The habitat of the early vertebrates. Biological Reviews,
32, 156-187. ROMER, A. S. 1966. Vertebrate paleontology. University of Chicago Press, Chicago,
468 pp. —— and PARSON, T. S. 1986. The vertebrate body. Saunders College Publishing,
Philadelphia, 679 pp. SHU, D.-G., CONWAY-MORRIS, S., HAN, J., ZHANG, Z.-F., YASUI, K.,
JANVIER, P., CHEN, L., ZHANG, X.-L., LIU, J.-N. and LIU, H.-Q. 2003. Head and backbone of the Early Cambrian vertebrate Haikouichthys. Nature, 431, 536-539.
—— LUO, H.-L., CONWAY-MORRIS, S., ZHANG, X.-L., HU, S.-X., CHEN, L., HAN, J., ZHU, M. and CHEN, L.-Z. 1999. Lower Cambrian vertebrates from South China. Nature, 402, 42-46.
SHUBIN, N., TABIN, C. and CARROLL, S. 1997. Fossils, genes and the evolution of animal limbs. Nature, 388, 639-648.
SMITH, I. C. 1957. New restorations of the heads of Pharyngolepis oblongus Kiaer and Pharyngolepis kiaeri sp. nov., with a note on their lateral line systems. Norsk Geologisk Tidsskrift, 37, 373-402.
SMITH, M. M. and HALL, B. K. 1990. Development and evolutionary origins of vertebrate skeletogenic and odontogenic tissues. Biological Reviews, 65, 277-373.
STENSIÖ, E. 1958. Les cyclostomes fossiles ou ostracoderms. 173-425. In GRASSÉ, P. (ed.). Traité de Zoologie, 13. Masson et Cie, Paris, 924 pp.
—— 1964. Les Cyclostomes fossiles ou ostracodermes. 96-382. In PIVETEAU, J. (ed.). Traite de Paleontologie, 4. Masson et Cie, Paris, 387 pp.
SWINTON, W. E. 1961. The story of prehistoric animals. Rathbone Books, London, 93 pp.
TARLO, L. B. 1960. The invertebrate origins of the vertebrates. Report of the 21st International Geological Congreess, Copenhagen, 22, 113-123.
WÄNGSJÖ, G. 1952. The Downtownian and Devonian vertebrates of Spitsbergen. IX Morphologic and systematic studies of the Spitsbergen Cephalaspids. Norsk Polarinstitutt Skrifter, 97, 1-611+addendum.
WESTOLL, T. S. 1958. The lateral fin-fold theory. 180-211. In WESTOLL, T. S. (ed.). Studies on fossil vertebrates. The Athelone Press, London, 263 pp.
WHITE, E. I. 1946. Jamoytius kerwoodi, a new chordate from the Silurian of Lanarkshire. Geological Magazine, 83, 89-97.
—— 1958. Original environment of the craniates. 212-234. In WESTOLL, T. S. (ed.). Studies on fossil vertebrates. The Athelone Press, London, 263 pp.
WICKSTEAD, J. M. 1969. Some further comments on Jamoytius kerwoodi White. Zoological Journal of the Linnean Society, 48, 421-479.
YOUNG, J. Z. 1981. The life of vertebrates. Clarendon Press, Oxford, 645 pp.
1
APPENDIX
Character list and matrix used in phylogenetic analysis – an updated version of Gess
et al. (2006) with neurological characters adapted according to P. Donoghue
(unpublished data). Changes to the matrix reflect changes due to 1) corrections of
coding strategy, 2) correcting coding to reflect new or updated character observations
and 3) new taxa and characters.
(a) Brain, sensory and nervous system
1. Neural crest absent = 0, present = 1
2. Olfactory peduncles absent = 0, present = 1
3. Pineal organ (extra-ocular photoreceptor region expressing pineal opsins)
absent = 0, present = 1
4. Adenohypophysis absent = 0, present = 1
5. Adenohypophysis simple = 0, compartmentalized = 1
6. Optic tectum absent = 0, present = 1
7. Cerebellar primordia absent = 0, present = 1
8. Pretrematic branches in branchial nerves absent = 0, present = 1
9. Flattened spinal chord absent = 0, present = 1
10. Ventral and dorsal spinal nerve roots united, absent = 0, present = 1
11. Mauthner fibres in central nervous system absent = 0, present = 1
12. Retina absent = 0, present = 1
13. Olfactory organ with external opening absent = 0, present = 1
14. Nasohypophyseal opening serving respiration (nasohypophyseal duct)
absent = 0, present = 1
15. Single nasohypophyseal opening, absent = 0, present = 1
16. Position of nasohypophyseal opening: terminal = 0, dorsal = 1
2
17. Olfactory organ unpaired = 0, paired = 1
18. Extrinsic eye musculature absent = 0, present = 1
19. Otic capsule anterior to branchial series, absent = 0, present = 1
20. Semicircular canals in labyrinth absent = 0, present = 1
21. Vertical semicircular canals forming loops, absent = 0, present = 1
22. Externally open endolymphatic ducts absent = 0, present = 1
23. Electroreceptive cells absent = 0, present = 1
24. Sensory lines absent = 0, present = 1
25. Sensory-lines on head only = 0, on head plus body = 1
26. Sensory-line enclosed in grooves = 0, enclosed in canals = 1
(b) Mouth and branchial system
27. Pouch-shaped gills absent = 0, present = 1
28. Single confluent branchial opening, absent = 0, present = 1
29. Elongate branchial series: more than 10 gill pouches/slits = 0, fewer than
10 = 1
30. Gill openings lateral and arranged in slanting row, absent = 0, present = 1
31. Position of gill openings: laterally = 0, ventrally = 1
32. Opercular flaps associated with gill openings, absent = 0, present = 1
33. Endodermal gill lamellae, absent = 0, present = 1
34. Gill lamellae with filaments, absent = 0, present = 1
35. Mouth terminal = 0, ventral = 1
36. Oral hood absent = 0, present = 1
37. Velum absent = 0, present = 1
3
(c) Circulatory system
38. Multi-chamber heart absent = 0, present = 1
39. Closed pericardium absent = 0, present = 1
40. Open blood system absent = 0, present = 1
41. Paired dorsal aortae absent = 0, present = 1
42. Large lateral head vein absent = 0, present = 1
43. Lymphocytes absent = 0, present = 1
44. Subaponeurotic vascular plexus absent = 0, present = 1
(d) Fins and fin-folds
45. Dorsal fin: separate dorsal fin absent = 0, present = 1
46. Dorsal fin originates at posterior of branchial series = 0, restricted to
posterior of trunk and/or caudal region = 1
47. Anal fin separate, absent = 0, present = 1
48. Fin ray supports, absent = 0, present = 1
49. Paired antero-posterior skin folds absent = 0, present = 1
50. Constricted pectoral fins with endoskeletal elements absent = 0, present =
1
51. Pelvic fins/flap, absent = 0, present = 1
52. Tail shape: no distinct lobes developed = 0, ventral lobe much larger than
dorsal = 1, dorsal lobe much larger than ventral = 2, dorsal and ventral
lobes almost equally developed = 3
53. Chordal disposition relative to tail development, isochordal = 0,
hypochordal = 1, hyperchordal = 2
54. Preanal median fold absent = 0, present = 1
4
(e) Skeletal
55. Ability to synthesise creatine phosphatase absent = 0, present = 1
56. Visceral arches fused to the neurocranium absent = 0, present =1
57. Keratinous teeth absent = 0, present = 1
58. Circumoral teeth absent = 0, present = 1
59. Circumoral teeth arranged in radiating series, absent = 0, present = 1
60. Trematic rings absent = 0, present = 1
61. Arcualia absent = 0, present = 1
62. Piston cartilage and apical plate, absent = 0, present = 1
63. Midline retractor muscle and paired protractor muscles, absent = 0,
present = 1
64. Transversely biting teeth (the wording of this character description has
been modified in order reduce ambiguity; coding reflects Gess et al.
2006) , absent = 0, present 1
65. Jaws (dorsoventral bite), absent 0, present = 1
66. Chondroitin 6-sulphate in cartilage, absent = 0, present = 1
67. Braincase with lateral walls, absent = 0, present = 1
68. Neurocranium entirely closed dorsally and covering the brain, absent = 0,
present = 1
69. Occiput enclosing vagus and glossopharyngeal nerves, absent = 0, present
= 1
70. Annular cartilage absent = 0, present = 1
71. Large oral disc absent = 0, present = 1
72. Tentacle cartilages; absent = 0, present = 1
5
73. Trunk dermal skeleton absent = 0, present = 1
74. Perichondral bone absent = 0, present = 1
75. Calcified cartilage absent = 0, present = 1
76. Cartilage composed of huge clumped chondrocytes, absent = 0, present =
1
77. Calcified dermal skeleton absent = 0, present = 1
78. Lamellar aspidin, absent = 0, present = 1
79. Cellular bone, absent = 0, present = 1
80. Dentine absent = 0, present = 1. Dentinous tissues are preserved in same
deposits as Jamoytius in the thelodont Loganellia, but there is no evidence
of dentine in any Jamoytius specimen. Jamoytius can therefore be reliably
interpreted as lacking dentine.
81. Dentine present as mesodentine = 0, orthodentine = 1
82. Enamel/oid absent = 0, (monotypic) enamel = 1, enameloid (bitypic
enamel) = 2
83. Three-layered exoskeleton consisting of a basal lamella, middle spongy
(or cancellar) layer and a superficial (often ornamented) layer: absent = 0,
present = 1
84. Cancellar layer in exoskeleton, with honeycomb-shaped cavities, absent =
0, present = 1
85. Scales/denticles/teeth composed of odontodes absent = 0, present = 1
86. Scale shape: diamond-shaped = 0, rod-shaped = 1
87. Oak-leaf-shaped tubercles, absent = 0, present = 1
88. Oral plates absent = 0, present = 1
89. Denticles in pharynx absent = 0, present = 1
6
90. Dermal head covering in adult state absent = 0, present = 1
91. Large unpaired ventral and dorsal dermal plates on head, absent = 0,
present = 1
92. Massive endoskeletal head shield covering the gills dorsally, absent = 0,
present = 1
93. Sclerotic ossicles absent = 0, present = 1
94. Ossified endoskeletal sclera encapsulating the eye, absent = 0, present = 1
(g) Miscellaneous
95. High blood pressure, absent = 0, present = 1
96. Hyperosmoregulation, absent = 0, present = 1
97. Male gametes shed directly through the coelom, absent = 0, present = 1
98. Forward migration of postotic myomeres, absent = 0, present = 1
99. Larval phase, absent = 0, present = 1
(h) Additional characters due to change in coding strategy and the previous
characters upon which they are contingent
100. (3.) Pineal opening covered = 0, uncovered = 1
101. (13.) External nasal opening single = 0, paired = 1
102. (20.) Number of semicircular canals one = 0, two = 1, three = 2
103. (24.) Neuromasts in sensory liness absent = 0, present = 1
104. (38.) Relative position of atrium and ventricle of heart: well separated = 0,
close to each other = 1
105. (43.) Lymphocytes antigen receptors VLR = 0, T and B = 1
7
106. (49.) Paired antero-posterior skin folds extend along the trunk = 0,
anterior only =1
107. (61.) Ventral arcualia absent = 0, present = 1
108. (85.) Scales/denticles/teeth made up by single odontode = 0, made up by
several odontodes = 1
109. (90.) Dermal head covering in adult state: micromeric = 0, large
(macromeric) dermal plates or shield = 1
Tunicata 1-10-0000-000-0---?0--00--010??00000001000000?0000000?0000-000-00-00000000000000--0-0-00-00000001011--------- Cephalochordata 0-10-00000000-0---?0--00--000000100110-010000?000000010000-000-00-00000000000000--0-0-00-00000000011--------- Myxinoidea 100101001101111000110001001*000010001100101101000000111110-1?1110000000100010000--0-0-00-0000000110-00000-?-- Myxinikela ????????????1?10??1???????101?0???00????????010?0000?0???0-?????0?????01000?0000--0-0-00-0-000??????0????-?-- Petromyzontida 10111110101110111111001110101100110111110010110100011011111111110110011000010000--0-0-00-0-00011111101111-0-- Mesomyzon ??????????????????1???????101?0??101?????????10?000110???11?????0?????10000?0000--0-0-00-0-000???????????-?-- Priscomyzon ??????????????????0???????101?0???011???????000?000?????1101????0????110000?0000--0-0-00-0-000???????????-?-- Mayomyzon ?????????????0????0???????101?0???011???????010?000*0??????1?1??0????100000?0000--0-0-00-0-000???????????-?-- Haikouichthys ??????????????????1????????01000??00????????00010000?1??00-?10-00?????00000?0000--0-0-00-0-000???????????-0-- Heterostraci ?11???1?????????1?0110?111111000?100???????101000003111??0-?1??00?????00100?110110111001011000?????0?1???-?11 Astraspis ??1??????????????????0?100101100???????????1?1??000???1?????????0???????100?11011210101??11??0?????0?????-?10 Arandaspida ??1????????????????????110100100??00????????010?000??01??0-?????0?????001???1101?2111111011?1??????0?????-?11 Anaspida ??1?????????1?11???????110100100??00????????01111001101??0-?????0?????00100?1100--0-1101010000?????10????*?10
8
Jamoytius ????????????1?10???????????0010???10????????????100??01???-?????0????1001???1??0--0-?100-0-000??????0????0??- Euphanerops ??????????????????????????10010??110?????????11110011?1?????1???0????100?0?10000--?-0?00-0-000???????????01-- Osteostraci ?011?11?????1011?10111?111101-11?110?11?01?1110101022011?0-?1??00?1110001110101100101001010111?????101?1?-?11 Galeaspida ?11??11?????11111?0111?111101-10?110?????1?1?1??000??011?0-?1??00?11100010101100--0-1001010100?????101???-?01 Loganellia ???????????????????????111?01101??00????????11101003101??0-????00?????00100?1101000-1000110000???????????1?00 Turinia ??????????????????????????1011?1??00?????????1?010??1?1?????????0?????0010??1101100-10001100?0???????????1?00 Jawedvertebrates 1111111101111-0-111111111100100101000111111011110112201000-010-011111000111010111110100011001111000012111-110 Euconodonta ?????????????????1?????????????????0????????01010001101?00-????10????00?000?1001?10-1?0010-000???????????-?1- Cornovichthys ??????????????????1?????????010?????????????0-0??00111??????????0???????000-0?00--??0-0?-0-0?0?????????????-- Achanarella ????????????????????????????0?0?????????????111??00?&???????????0???????000-0?00--??0-0?-0-0???????????????-- *=0/1 &=1/2