Phylogenetic analysis and revision of the trilobite subfamily Balnibarbiinae (Olenidae)

AMERICAN MUSEUM NOVITATES

Number 3928, 20 pp.

May 7,2019

Phylogenetic analysis and revision of the trilobite
subfamily Balnibarbiinae (Olenidae)

MELANIE J. HOPKINS'

ABSTRACT

The Balnibarbiinae is one of eight subfamilies of the Olenidae, a diverse family of late
Cambrian to Ordovician trilobites. Balnibarbiine species occur in a relatively continuous
section of deeper-water sediments exposed along the northeastern coastline of Spitsbergen,
Svalbard, as well as scattered deeper-water beds in central Nevada. Results of phylogenetic
analyses of the subfamily using both parsimony and Bayesian methods are consistent with a
previous hypothesis based on phyletic similarity and stratigraphic range. Cloacaspis Fortey,
1974, is supported as monophyletic, but the support for Balniharbi Fortey, 1974, is weak, and
the genus may be paraphyletic to Cloacaspis even with the reassignment of Balniharbi ceryx
Fortey, 1974, to Cloacaspis. New field collections and discovery of previously undescribed
material in museum and survey collections provides the basis for emended descriptions of
the genus Cloacaspis, as well as Cloacaspis tesselata Fortey and Droser, 1999, Cloacaspis ekphy-
mosa Fortey, 1974, and Balniharbi erugata Fortey, 1974, and expands the geographic range
of the subfamily to Alaska.

INTRODUCTION

The Olenidae is a diverse family of trilobites (408 species in 68 genera; Adrain, 2011) rang¬
ing from the Guzhangian to the end of the Ordovician (Adrain, 2013). The Balnibarbiinae is
one of eight subfamilies within Olenidae, and is known almost exclusively from deeper-water
deposits in Ny Friesland, northeastern Spitsbergen, Svalbard (Fortey, 1974; Kroger et al., 2017).
The pre-Carboniferous basement of the Svalbard archipelago consists of several tectonostrati-

^ Division of Paleontology, American Museum of Natural History
Copyright ® American Museum of Natural History 2019

ISSN 0003-0082

2

AMERICAN MUSEUM NOVITATES

NO. 3928

graphically distinct terranes that were stretched along the margin of Laurentia in pre-Caledo¬
nian times (Gee and Page, 1994; Gee and Tebenkov, 2004). Other balnibarbiine occurrences
are in deeper-water deposits in Nevada and Alaska (Ethington et al., 1995; Fortey and Droser,
1999; this study). Balnibarbiine specimens are much rarer in western Laurentia, however, pri¬
marily because deeper-water sediments are rare in the Early to Middle Ordovician rock record
preserved there. In contrast, the section at Ny Friesland is one of the most continuous deeper-
water sections of Early to Middle Ordovician age anywhere in the world.

The subfamily Balnibarbiinae was first described based on collections made during
two expeditions to Ny Friesland, Spitsbergen, in 1967 (Vallance and Fortey, 1968) and 1972
(Fortey and Bruton, 1973; see also Fortey and Bruton, 2013). The abundance of specimens
in the Spitsbergen sections made it possible to infer changes in the exoskeletal morphology
through stratigraphic time. Based on such observations, Fortey (1974) proposed a phylog-
eny consisting of three evolutionary lineages. In his conception, the genus Balnibarbi
Fortey, 1974, comprised two basal evolutionary lineages and was paraphyletic to the third
evolutionary lineage, consisting of species of the genus Cloacaspis Fortey, 1974. During
this time, undescribed olenids were also reported from late Early to early Middle Ordovi¬
cian deposits in Nevada (McKee et al., 1972), but it was not until the 1990s that these were
recognized as having an affinity with Spitsbergen balnibarbiines (Ethington et al., 1995;
Fortey and Droser, 1999).

Recently, the author collected new balnibarbiine specimens from both Nevada (2015) and
Spitsbergen (2016). The purpose of this study is to use new collections and modern methods
to revise the subfamily and test Fortey s (1974) phylogenetic hypothesis.

PHYLOGENETIC ANALYSIS
Materials and Methods

Character design and coding: Forty-five characters are included in the analysis, of
which 32 describe the cranidium, four describe the librigenae, and nine describe the pygidium.
Characters describing the thorax and hypostome are excluded because these sclerites are
unknown for most taxa in the analysis. Characters were coded using reductive coding sensu
Strong and Lipscomb (1999). All taxa (including outgroup species, see below) share the fol¬
lowing traits: curved and divergent anterior facial suture, deflected SI, and occipital node,
where known. These three traits vary in how commonly they occur among olenids (e.g., Monti
and Confalonieri, 2019), but because they are constant among the taxa in this analysis, they
are excluded from the character matrix. The absence or presence of triangular pleural nodes is
also excluded because, while this trait is likely shared by all balnibarbiine species to the exclu¬
sion of other olenid species, its presence or absence cannot be coded for the outgroup taxa (for
which pygidia are unknown). The data matrix is archived in MorphoBank (http://morphobank.
org/permalink/?P3234).

Fortey and Droser (1999) did not describe the pygidium for Cloacaspis tesselata. Based
on association of specimens, a pygidium found in the U.S. Geological Survey collections is

2019

HOPKINS: THE TRILOBITE SUBEAMILY BALNIBARBIINAE

3

tentatively assigned to this species (see Systematic Paleontology). The full character matrix
includes character states coded from this specimen, but the analysis was also rerun, treating
these states as missing.

Outgroup selection: Fortey (1974) noted that the glabellae of some Parabolinella Brog-
ger, 1882, species resembled those of balnibarbiine species. Parabolinella prolata Robison and
Pantoja-Alor, 1968, and Parabolinella tumifrons Kobayashi, 1936, were selected as outgroup
taxa because they have Laurentian occurrences as well as occurrences during the early Skull-
rockian (i.e., they are older than the ingroup taxa). They are also placed basal to other Parabo¬
linella species in recent phylogenetic analyses (Monti and Confalonieri, 2013, 2017). Fortey
(1974) also suggested that Agalatus {= Inkouia, fide Zhang, 1985) was similar to Balnibarbi and
CloacaspiSy but excluded the genus from the Balnibarbiinae because species lacked the diag¬
nostic triangular pleural node. Although this overall similarity makes the genus a candidate
outgroup, no Inkuoia species were included in the outgroup because it was not possible to code
most of the characters from available figures showing specimens assigned to Inkouia species
(e.g., Lisogor, 1961; Han, 1983).

Tree searching: Heuristic searching in PAUP'^4.0bl0 (Swofford, 1998) was used to find
the optimal tree according to the maximum parsimony criterion. Inapplicable characters
were treated as missing data. Taxa were added by random sequence addition with 100 rep¬
licates and branch swapping was performed using the tree bisection reconnection option
(TBR). Five characters (1, 9, 14, 19, 33) describing continuous variation were treated as
ordered. All characters were weighted equally. Clade support was assessed based on Bremer
support values, and bootstrap and jackknife analyses (with 33% deletion of characters), each
consisting of 1000 replicates.

For comparison, a Bayesian search was conducted in MrBayes 3.2.6 (Ronquist et al., 2012),
employing the Mk model (Lewis, 2001), specifying that only variable characters were sampled,
assuming no rate variation across characters, and assigning Parabolinella prolata as outgroup.
Different analyses were run assuming no character rate variation, and both gamma- and log-
normally {K = 4) distributed character rate variation (Harrison and Larsson, 2015); 500,000
MCMC repetitions were required for convergence for all analyses.

Results

The parsimony-based tree search recovered three most parsimonious trees with tree
length of 98. All nodes are resolved in the strict consensus tree (fig. 1) except those leading
to the clade comprising Cloacaspis tesselata, Clocacaspis senilis^ and Cloacaspis ekphymosa.
In the Bayesian analysis, a node uniting C. senilis and C. ekphymosa is recovered with a
posterior probability of 55 regardless of settings for character rate variation (not shown).
In both analyses, the two Cloacapsis ceryx subspecies are sister to one another, although
this node is not as well supported in the parsimony tree as that uniting more-derived
Cloacaspis species or the node uniting all Cloacaspis species, and is recovered in some
maximum clade credibility trees but with posterior probability <50. Balnibarbi pulvurea

AMERICAN MUSEUM NOVITATES

NO. 3928

A

B

c

CD

CO

.CO

2019

HOPKINS: THE TRILOBITE SUBEAMILY BALNIBARBIINAE

5

and Balnibarbi tholia are recovered as sister taxa, and Balnibarbi erugata subspecies are
supported as sister taxa in both analyses with moderate support. Balnibarbi is weakly sup¬
ported as monophyletic in the parsimony tree and in the maximum clade credibility tree
assuming no rate variation. If log-normal or gamma-shaped rate variation across charac¬
ters is assumed instead of equal rates, maximum clade credibility trees indicate weak sup¬
port for a paraphyletic Balnibarbi: the branch leading to B. tholia and B. pulvurea attaches
below the node uniting Cloacaspis. For all Bayesian analyses, however, nodal support is
low (posterior probabilities <50), so these relationships are effectively unresolved at this
point. Although it is possible that Balnibarbi is actually paraphyletic to Cloacaspis, as con¬
ceived by Fortey (1974), all species except C. ceryx are retained within Balnibarbi. Coding
pygidial characters as missing for Cloacaspis tesselata did not change the tree topologies
or node support.

SYSTEMATIC PALEONTOLOGY

Specimens were examined from collections in the Natural History Museum, Oslo (PMO-
NE), Sedgwick Museum, University of Cambridge (CAMSM), Cambridge Arctic Shelf Pro¬
gramme (CASP), the US. Geological Survey (USGS), the Smithsonian (USNM), and the
American Museum of Natural History (AMNH). Because there were differences in section
measurements taken between Spitsbergen expeditions (compare Fortey, 1980, with Kroger et
al., 2017), stratigraphic ranges in figure 1 are shown scaled to the 2016 sections and include
range extensions based on new collections (see below).

■< -

EIGURE 1. Results of phylogenetic analysis. A. Stratigraphic section of Olenidsletta Member, Profilstranda,
Ny Friesland, Spitsbergen. Modified from Kroger et al., 2017. B. Strict consensus tree of three most parsimoni¬
ous trees, scaled to time for graphical representation purposes using the all-branches additive-method (nodes
scaled to first occurrence of oldest descendent and an arbitrary constant value is added to all branches in order
to remove zero-length branches). Maximum clade credibility tree resulting from Bayesian analysis assuming
no rate variation is the same as the strict consensus tree except that the node joining C. senilis and C. ekphy-
mosa is also recovered with a posterior probability of 55. The only difference in the topology of the maximum
clade credibility tree if rates are assumed to be gamma- or log-normally distributed is that the branch leading
to B. tholia and B. pulvurea attaches below the node uniting Cloacaspis, which is also more strongly supported
(posterior probability = 70); 50% majority-rule consensus trees are unable to resolve Balnibarbi as a mono¬
phyletic clade (all nodes with posterior probabilities <50 collapse). Values to left of nodes show Bremer sup¬
port values/bootstrap frequencies/jackknife frequencies. Italicized values to right of nodes show posterior
probabilities for analyses assuming no rate variation/gamma-distributed rate variation/log-normally distrib¬
uted rate variation; NA indicates that the node shown was not supported in the maximum clade credibility
tree for that analysis. Stratigraphic ranges based on Fortey (1980), Fortey and Droser (1999), and new collec¬
tions that extend the range of Cloacaspis ceryx anataphra and Balnibarbi scimitar. Cloacaspis tesselata is Ran-
gerian and likely just younger than C. senilis.

6

AMERICAN MUSEUM NOVITATES

NO. 3928

CLASS TRILOBITA WALSH, 1771
ORDER OLENIDA ADRAIN, 2011
EAMILY OLENIDAE BURMEISTER, 1843
SUBEAMILY BALNIBARBllNAE EORTEY, 1974
Balniharhi Eortey, 1974

Type species: Balnibarbi pulvurea Eortey, 1974.

Diagnosis: As in Eortey, 1974: 21.

Included species: Balnibarbi pulvurea Eortey, 1974; Balnibarbi erugata erugata Eortey,
1974; Balnibarbi erugata sombrero (Eortey, 1974); Balnibarbi scimitar Eortey, 1974; Balnibarbi
tholia Eortey, 1974.

Balnibarbi erugata Eortey, 1974
Eigure 2

Type subspecies: Balnibarbi erugata erugata Eortey, 1974.

Type locality: Profilstranda, Olenidsletta Member, Valhallfonna Eormation, Ny Eriesland,
Spitsbergen, Svalbard.

Subspecies: Balnibarbi erugata erugata Eortey, 1974, Balnibarbi erugata sombrero (Eortey,
1974).

Emended diagnosis: Balnibarbi species with long (sag.) anterior border rounded on mid¬
line. Anterior border with 25-30 pits; pits expressed only on internal mold. Glabella rounded
anteriorly; preglabellar field broad. Palpebral lobes long, with anterior end forward of 2p and
posterior end at occipital furrow. Moderately large pygidium with four axial rings.

Discussion: In both subspecies, distinct pits are visible only on the internal mold of the
anterior border furrow (fig. 2E, E). In this case (where the pit is not expressed on the dorsal
surface), this trait may be better described as a series of “protuberances” on the ventral surface
that form pits on the internal mold. These protuberances meet nodes on the doublure of the
librigena beneath the anterior border of the cranidium (see Eortey, 1974: pi. 5, fig. 10). On other
Balnibarbi species, the structures are visible on the anterior border (e.g., Balnibarbi pulvurea,
see Eortey, 1974: pi. 1, fig. 4), and more in keeping with the idea of a “pit.”

The holotype of Balnibarbi erugata erugata is almost completely exfoliated (fig. 2A),
but specimens that otherwise fit the diagnosis (including some listed by Eortey, 1974) show
very fine granulation across the glabella (fig. 2D), in the posterior border furrow (fig. 2C),
and more rarely preserved on the frontal area (fig. 2B). Terrace lines are evident on the
border of the holotype (fig. 2A) and some better-preserved specimens (fig. 2B). Eortey
(1974) differentiated Balnibarbi erugata and Balnibarbi sombrero based on the length (sag.)
of the preglabellar field relative to the glabella. The type and figured specimens of B. som¬
brero do have a larger preglabellar field (fig. 3), and there is no granulation visible on the
external surface where preserved. However, some specimens that have fine granulation

2019

HOPKINS: THE TRILOBITE SUBEAMILY BALNIBARBIINAE

7

EIGURE 2. Balnibarbi erugata Fortey, 1974, close-ups. A. Balnibarbi erugata erugata, holotype cranidium,
PMO-NF-3016. Note exfoliation of frontal area and glabella. Arrow points to terrace lines on anterior border
that are expressed on both the external surface and internal mold. B. Anterior border of Balnibarbi erugata
erugata, CAMSM F3976, showing faint caeca and very fine granulation on frontal area near anterior border
furrow and terrace lines on border. C. Posterior border furrow on Balnibarbi erugata erugata, CAMSM F3976,
showing very fine granulation. D. Occipital ring of Balnibarbi erugata erugata, PMO-NF-695, showing very
fine granulation on glabella. E. Balnibarbi erugata erugata PMO-NF-772. Arrows point to select pits in ante¬
rior border furrow; specimen is exfoliated. F. Balnibarbi erugata sombrero, PMO-NF-2054. Arrows point to
select pits in anterior border furrow on exfoliated surface. Scale bars = 1 mm.

AMERICAN MUSEUM NOVITATES

NO. 3928

also have large preglabellar fields (e.g., PMO-NF-227.730). In addition, fine granulation is
easily eroded away on exposed surfaces (see fig. 4F). No new specimens of Balnibarbi
erugata sombrero were recovered during the 2016 expedition to Spitsbergen, nor were any
Balnibarbi erugata erugata specimens collected from trilobite zone Vlb, thus the known
stratigraphic ranges are still nonoverlapping. This separation in stratigraphic time may
have contributed to Fortey s decision to rank each at the species level. However, the two
subspecies differ morphologically in the same way that subspecies of Cloacaspis ceryx do
(expression of granulation and morphometric differences in the frontal area). Thus, B.
sombrero was lowered to subspecies status; the name erugata takes precedence following
ICZN Article 24.2.

Balnibarbi erugata erugata Fortey, 1974
Figure 2A-D, E

Balnibarbi erugata Fortey, 1974: 31, pi. 5, fig. 1-10.

Holotype: Cranidium, PMO NF 3016 (figured in Fortey, 1974: pi. 5, fig. 1-3).

Type locality: Profilstranda, Olenidsletta Member, Valhallfonna Formation, Ny Friesland,
Spitsbergen, Svalbard.

Stratigraphic range: Throughout trilobite zone Vic and the very beginning of V2a
(Fortey, 1980). These zones coincide with the upper Oepikodus evae and lower Oepikodus inter-
medius conodont zones, and the Didymographtus protobifidus graptolite zone within the upper
Floian (fig. 1; see also Krdger et al., 2017).

Diagnosis: Balnibarbi erugata subspecies with relatively short (sag.) preglabellar field
(<0.35 length of glabella) and fine granulation present on glabella, axial, palpebral, and poste¬
rior border furrows, and on frontal area adjacent to anterior border furrow.

Balnibarbi erugata sombrero (Fortey, 1974)

Figure 2F

Balnibarbi sombrero Fortey, 1974: 32, pi. 6, fig. 1-4.

Holotype: Cranidium, PMO NF 835 (figured in Fortey, 1974: pi. 6, fig. 1-2).

Type locality: Profilstranda, Olenidsletta Member, Valhallfonna Formation, Ny Friesland,
Spitsbergen, Svalbard.

Stratigraphic range: Middle of trilobite zone Vlb (Fortey, 1980). This zone coincides
with the lower Oepikodus evae conodont zone and Pendeograptus fruticosus graptolite zone
within the upper Floian (fig. 1; see also Krdger et al., 2017).

Diagnosis: Balnibarbi erugata species with relatively long (sag.) preglabellar field (>0.35
length of glabella) and smooth exoskeletal surface. Known only from cranidia.

2019

HOPKINS: THE TRILOBITE SUBEAMILY BALNIBARBIINAE

9

X
I—
CD
Z
LJJ
_l

<

_i

_i

LJJ

CO

<

_l

X
I—
CD
Z
UJ
_l

<

_l

_l

UJ

CQ

CD

UJ

X

Q_

A

A

• B. erugata sombrero

• B. erugata erugata
^ C. ceryx ceryx

^ C. ceryx anataphra

5

10

15

20

CRANIDIAL LENGTH (MM)

EIGURE 3. Size of preglabellar field relative to glabella in subspecies of Balnibarbi erugata and Cloacaspis
ceryx. Arrow points to PMO-NF-227.730. Differences in relative length of preglabellar field are not correlated
with overall size of the cranidium, and are thus not due to allometry. However, while subspecies of Balnibarbi
erugata overlap in cranidial length, all measured specimens of C. ceryx anataphra are smaller than those of
C. ceryx ceryx. Since these subspecies also differ in expression of surface granulation, it is possible that expres¬
sion of granulation is related to size.

Balnibarbi scimitar, Fortey, 1974

Balnibarbi scimitar Fortey, 1974: 33, pi. 7, fig. 1-10.

Holotype: Cranidium, PMO NF 2785 (figured in Fortey, 1974: pi. 7, figs. 1, 3, 8).

Type locality: Melt stream D on Olenidsletta, Olenidsletta Member, Valhallfonna Forma¬
tion, Ny Friesland, Spitsbergen, Svalbard.

10

AMERICAN MUSEUM NOVITATES

NO. 3928

Stratigraphic range: The type material for this species was sampled from a single bed
estimated to be no more than 6 m from the base of the Olenidsletta member (Fortey, 1974),
which would put the occurrence in trilobite zone Via (Fortey, 1980; coincident with conodont
zone Oepikodus communis; see Kroger et al., 2017). New specimens were recovered in 2016
from the Profilstranda section from a bed 30 m above the base of the Olenidsletta member
which extends the stratigraphic range into trilobite zone Vlb (fig. 1).

Diagnosis: As in Fortey, 1974: 33.

Cloacaspis Fortey, 1974

Type species: Cloacaspis senilis Fortey, 1974.

Emended Diagnosis: Balnibarbiine trilobites with facial sutures moderately divergent in
front of eyes. Pits present on internal mold of anterior border furrow, and may also expressed
on the dorsal surface. Frontal area relatively short (sag.), ranging from 10%-20% of the total
cranidial sagittal length. Glabella with four pairs of glabellar furrows of balnibarbiinae type: IP
extends posterolaterally from axial furrow before turning abruptly posteriorly; 2P straight,
extending posterolaterally from axial furrow, or curving slightly posteriorly; 3P straight, short,
transverse, not reaching axial furrow; 4P straight, short, slightly oblique, not reaching axial
furrow. 3P and 4P tend to be longer than in other balnibarbiine genera {Balnibarbi species),
and 4P is more weakly expressed than other furrows. Postocular fixed cheeks triangular. Pos¬
terior border of free cheek curves forward to long genal spine. Pygidium small, with two to
three axial rings. Posterior border of pygidium may bear short spines.

Included species: Cloacaspis senilis Fortey, 1974; Cloacaspis ceryx (Fortey, 1974); Cloacas¬
pis dejecta Fortey, 1974; Cloacaspis ekphymosa Fortey, 1974; Cloacaspis tesselata Fortey and
Droser, 1999.

Cloacaspis ekphymosa Fortey, 1974
Figure 4A

Triarthrus sp. Ross, 1965: 18, pi. 8, fig. 4.

Cloacaspis ekphymosa Fortey, 1974: 41, pi. 11, figs. 12, 14-18.

Holotype: Cranidium, SMA 84075 (figured in Fortey, 1974: pi. 11, fig. 12).

Type locality: Profilstranda, Olenidsletta Member, Valhallfonna Formation, Ny Friesland,
Spitsbergen, Svalbard.

Other geographic occurrences: Lost River Area, Seward Peninsula, Alaska (see Ross,
1965).

Stratigraphic range: Fortey (1974, 1980) recovered this species from 105 m to 120 m
in the section of Fortey (1980), stratigraphically below the congeneric Cloacaspis senilis, which
was recovered from 120 m to 145 m above the base of the Olenidsletta Member. This species
was recovered in 2016 cooccurring with Cloacaspis senilis at 139 m above the base of the

2019

HOPKINS; THE TRILOBITE SUBEAMILY BALNIBARBIINAE

11

EIGURE 4. Cloacaspis spp. A. Cloacaspis ekphymosa Fortey, 1974, USNM 144232. Collected from the Lost
River Area on Seward Peninsula, Alaska, and previously identified as Triarthrus sp. (Ross, 1965). Arrow point¬
ing to postpalpebral furrow. B. Anterior border of Cloacaspis ceryx anataphra (Fortey, 1974), AMNH-
FI-101655. Black arrows pointing to faint anterior border pits on internal mold. C. Cloacaspis ceryx anataphra
(Fortey, 1974), AMNH-FI-101655. Collected from PO-18 of the Olenidsletta Member of the Valhallfonna
Formation (Kroger et al., 2017). D) Cloacaspis ceryx ceryx (Fortey, 1974), AMNH-FI-101657. Close-up of E,
anterior glabellar lobe and F, posterior wing, showing presence of fine granulation of recently prepared surface
(white arrow) and lack of granulation on exposed and slightly eroded surface (black arrow). Collected from
bed PO-18 of the Olenidsletta Member of the Valhallfonna Formation (Kroger et al., 2017). Scale bars = 1
mm except A and D, where scale bar = 5 mm.

12

AMERICAN MUSEUM NOVITATES

NO. 3928

Olenidsletta Member at the Profilstranda section (Kroger et al., 2017; see also fig. 1), which is
estimated to be at about 119 m within the stratigraphic section of Fortey (1980). It is thus not
clear whether this should be considered an extension of the stratigraphic range of Cloacaspis
senilis or of Cloacaspis ekphymosa, but does indicate that the species overlapped just after the
first appearance of the former and just before the last appearance of the latter.

Diagnosis: As in Fortey, 1974: 41-42.

Discussion: Fortey (1974: 14) suggested that the ''Triarthrus' cranidium figured by Ross
(1965: pi. 8, fig. 4) was more likely Cloacaspis. Examination of this specimen confirms this, and
the presence of the postpalpebral furrow indicates that the specimen belongs to Cloacaspis
ekphymosa Fortey, 1974, although the extensive exfoliation does not allow for the confirmation
of granulate surface sculpture (fig. 4A). This identification extends the geographic range of this
species and genus to Alaska.

Cloacaspis ceryx (Fortey, 1974)

Figure 4B-F

Balnibarhi ceryx Fortey, 1974: 29.

Holotype: Cranidium, SMA 84034 (figured in Fortey, 1974: pi. 8, fig. 1).

Type locality: Profilstranda, Olenidsletta Member, Valhallfonna Formation, Ny Friesland,
Spitsbergen, Svalbard.

Emended Diagnosis: Cloacaspis species with greater (sag., exsag.) preglabellar length than
other species of the genus; anterior border comes to a point on midline. Moderately sized
pygidium with three axial rings.

Discussion: Eortey (1974) distinguished Cloacaspis ceryx ceryx and Cloacaspis ceryx anata-
phra by the latter having a shorter (sag.) preglabellar field relative to the glabella, no visible pits
in the anterior border furrow, and a smooth exoskeletal surface. Examination of new and old
specimens suggests that the mean preglabellar/glabella ratio is on average shorter in C. ceryx
anataphra, but there is overlap in variation in this variable. In addition, while pits are not visible
in the anterior border furrow, they are present and visible on the internal mold (fig. 4B, C). On
the few specimens where it was possible to estimate the number of anterior pits, the number is
slightly smaller (25-30) compared to C. ceryx ceryx (>30). Although granulation is very fine on
C. ceryx ceryx and its preservation is very sensitive to any erosion of the surface (fig. 4D-E), it
does appear that some specimens truly lack surface granulation. Based on this evidence, the two
taxa are retained as subspecies, but the species complex is reassigned to the genus Cloacaspis
based on the results of the phylogenetic analysis. The species complex is united with other Clo¬
acaspis species by having advanced genal spines (char. 34), relatively large palpebral lobes (char.
19), and forward curvature of the palpebral lobe (char. 20). However, Cloacaspis ceryx shares
some characters with some Balnibarhi species, including greater divergence of the anterior dorsal
suture (char. 9), relatively wide palpebral lobes (char. 21), relatively short 3P (char. 25) larger
number of axial rings (char. 37), and medial indent on posterior margin of pygidium (char. 44).
The latter three characters are also shared with Cloacaspis dejecta.

2019

HOPKINS; THE TRILOBITE SUBEAMILY BALNIBARBIINAE

13

Cloacaspis ceryx ceryx (Fortey, 1974)

Figure 4D-F

Balnibarhi ceryx ceryx Fortey, 1974: 28, pi. 8, fig. 1-6, pi. 9, figs. 1, 3, 4.

Holotype: Cranidium, SMA 84034 (figured in Fortey, 1974: pi. 8, fig. 1).

Type locality: Profilstranda, Olenidsletta Member, Valhallfonna Formation, Ny Friesland,
Spitsbergen, Svalbard.

Stratigraphic range: Lower Vlb trilobite zone (Fortey, 1980), which coincides with the
lower Oepikodus evae conodont zone and Pendograptus fruticosus zone within the upper Floian
(fig. 1; see also Kroger et al., 2017).

Diagnosis: Cloacaspis ceryx species with relatively long (sag.) preglabellar field (0.08 - 0.19
length of glabella) and finely granular exoskeletal surface.

Cloacaspis ceryx anataphra (Fortey, 1974)

Figure 4B, C

Balnibarhi ceryx anataphra Fortey, 1974: 30, pi. 9, fig. 5-7.

Holotype: Cranidium, PMO NF 651 (figured in Fortey, 1974: pi. 9, fig. 5).

Type locality: Profilstranda, Olenidsletta Member, Valhallfonna Formation, Ny Friesland,
Spitsbergen, Svalbard.

Stratigraphic range: Fortey (1974) reported this subspecies from a single horizon
about 8 m above the base of the Olenidsletta Member, which is estimated to be around 12.5
m in the 2016 stratigraphic section. Additional specimens were recovered in 2016 from
17-19 m in section, which extends the stratigraphic range of this subspecies, but the total
range remains within the early half of the stratigraphic range of the other subspecies, Clo¬
acaspis ceryx ceryx (fig. 1).

Diagnosis: Cloacaspis ceryx species with relatively narrow preglabellar field (0.06-0.10
length of glabella) and smooth exoskeletal surface.

Cloacaspis tesselata Fortey and Droser, 1999
Figures 5-7

Clocaspis tesselata Fortey and Droser, 1999: 187, fig. 3.1-3.5.

Holotype: Cranidium, USNM 495868 (figured in Fortey and Droser, 1999: fig. 3.4).
Type locality: “Olenid bed,” Antelope Valley Formation, Little Rawhide Mountain, Hot
Creek Range, Nye County, Nevada.

Other Occurrences: Antelope Valley Limestone, June Canyon Sequence, Ike’s Canyon,
Nevada.

Stratigraphic range: Lower Dapingian. See discussion for more detail.

14

AMERICAN MUSEUM NOVITATES

NO. 3928

EIGURE 5. Cloacaspis tesselata Fortey and Droser, 1999, specimens from Little Rawhide Mountain, Nevada.
A. Librigena, AMNH-FI-115819. Specimen oriented relative to cranidium shown in B. B. Cranidium, AMNH-
FI-115818. C. Librigena, AMNH-FI-115820. D. Paratype librigena, USNM 495690. Close-ups of B; E, poste¬
rior wing showing anastomizing lines along dorsal suture and glabella (white arrows) and curvature of distal
part of posterior furrow (black arrow); and F, glabellar lobe LI, showing anastomizing lines and very fine
granulation. G. Cranidium, AMNH-FI-115807. H. Cranidium, AMNH-FI-115808. All specimens except para¬
type collected by author in 2015. Scale bar = 1 mm.

2019

HOPKINS: THE TRILOBITE SUBEAMILY BALNIBARBIINAE

15

EIGURE 6. Cloacaspis tesselata Fortey and Droser, 1999, specimens from Ikes Canyon, Nevada. A. External
mold of cranidium, USNM 720100. From USGS-D2219. B. Latex mold of cranidium, AMNH-FI-115823, cast
from USNM 720100 (see A). C. Cranidium, AMNH-FI-115825. Found in Columbia University collections.
D. Cranidium, AMNH-FI-115824. Found in Columbia University collections. E. Cranidium and thorax,
USNM 720101. From USGS D2282-CO. Scale bar = 1 mm.

16

AMERICAN MUSEUM NOVITATES

NO. 3928

EIGURE 7. Cloacaspis tesselata? Fortey and Droser, 1999, specimens from Ikes Canyon, Nevada. A. Dorsal
view of fused librigena, USNM 720102 (part). From USGS-D2217. B. Ventral view of fused librigena, USNM
720102 (counterpart). C. Close-up of anterior part of librigena USNM 720102 (counterpart), showing fusion
and triangular projection. D. Close-up of advanced genal spine on librigena USNM 720102 (part). E. Pygid-
ium, USNM 720103. From USGS D2217. F. Lateral view of pygidium, USNM 720103. Scale bar = 1 mm,
except for A and B, where scale bar = 5 mm.

Diagnosis: As in Fortey and Droser, 1999: 187.

Emended Description: Posterior area of fixigenae broadly triangular in shape, steeply
downsloping. The posterior border is elevated to form an articulating socket close to the occipi¬
tal ring, as was described for Cloacaspis ceryx ceryx (Fortey, 1974) and is apparent in other
Cloacaspis species (Fortey, 1974). Fine anastomizing lines distributed across entire glabella, not
just frontal glabellar lobe; fine anastomizing lines on genal field bordering posterior dorsal
suture (fig. 5D, E). Very fine granulation apparent on glabella (fig. 5F). Occipital lobe has very
small medial tubercle. Librigenae fused at midline, with advanced genal spine, such that pos¬
terior border curves toward it. Triangular projection of anterior border at midline. Raised ridge
follows outline of eye, more prominent anteriorly (fig. 5A, C, E). Tentatively assigned pygidium
1.6 times as wide as long, with two axial rings and rounded terminal piece. Anterolateral mar¬
gin curves posteriorly from axis, posterior border smoothly curving. Pleural field slopping
ventrally, with two strongly expressed pleural furrows, two moderately strongly expressed inter¬
pleural furrows, and weakly visible triangular pleural nodes. No border furrow. Surface sculp¬
ture unknown (fig. 7E, F).

2019

HOPKINS: THE TRILOBITE SUBEAMILY BALNIBARBIINAE

17

Discussion: Cloacaspis tesselata was first described from the Antelope Valley Limestone
exposed at Little Rawhide Mountain, Nevada. Specimens are rare at this locality, having been
sparsely collected from a single 30 cm black limestone bed by Fortey and Droser in the mid- to
late-1990s, and by the author in 2015. A similar olenid had been reported from the Antelope
Valley Limestone in Ikes Canyon in the Toquima Range, Nye County, Nevada (McKee et al.,
1972; McKee, 1976; Ethington et al., 1995). The author found one scrappy olenid specimen
(AMNH-Fl-101458, not figured) over four days of fieldwork in Ike’s Canyon in 2015; however,
additional specimens were found in the Columbia University collections now housed at the
American Museum of Natural History (AMNH-Fl-115824-825) and in the U.S. Geological
Survey collections (USGS D2217-CO, D2219-CO, D2220-CO, D2280-CO, D2282-CO). Where
the glabellar furrows are adequately preserved, all cranidia have the distinctive bifurcating
glabellar furrows diagnostic of Cloacaspis tesselata; thus, it seems likely that other Cloacaspis-
like sclerites found in these collections also belong to Cloacaspis tesselata. Specimens include
complete fused librigena and a single pygidium (figs. 5, 6, 7), thereby making it possible to
expand on the description.

None of the paratype cranidia of Cloacaspis tesselata (USNM 495686-88) have the posterior
area of the fixigenae preserved. However, two new specimens, one collected from Little Rawhide
Mountain (fig. 5B) and one found in the USGS collections, show the entire dorsal suture includ¬
ing the posterior area of the fixigenae (fig. 6A-B). Comparisons of these cranidia with new libri-
genal specimens (figs. 5A, C, 7A, B, D) show that there is an advanced genal spine, in contrast to
Fortey and Droser’s (1999) description. It is also possible to see the distal end of the posterior
border furrow start to gently curve up on the posterior wing of the cranidium (fig. 5A, C). The
librigena assigned by Fortey and Droser (1999) appears to belong to Cloacaspis tesselata as well,
but the posterior margin is not preserved well enough to see the genal angle (fig. 5D). Other
librigenae from the USGS collections (fig. 7A-C) also show advanced genal spines (fig. 7D). Of
the specimens examined, the acuteness of the angle in the Ikes Canyon specimens is greater than
that of the specimens recovered from Little Rawhide Mountain (compare fig. 5A with fig. 7A, B,
and D), but it is possible that some of this variation is taphonomic, as the specimens from Ike’s
Canyon are flattened relative to the specimens from Little Rawhide Mountain. Some librigenae
recovered from Ike’s Canyon are fused (fig. 7A, B), as seen in other Cloacaspis species (and other
olenid trilobites), but there is also a triangular projection at the anterior midline of the fused
librigenae (fig. 7C) that has not been reported before. Fortey and Droser (1999) did not identify
any pygidia, and only one specimen was found among the USGS or AMNH collections (fig. 7E)
that could belong to Cloacaspis tesselata based on similarities to those described for Cloacaspis
dejecta (Fortey, 1974: pi. 12, figs. 1,4). The specimen is almost entirely exfoliated, so any surface
sculpture remains unknown, though it is possible that there is fine granulation on the bit of exo¬
skeleton on the right pleural region. One partially complete olenid specimen was found in USGS
collection D2282-CO. Triangular pleural nodes are preserved on the thorax, indicating that it
belongs to Balnibarbiinae. The glabellar furrows are poorly preserved, but the shape of the cra¬
nidium is consistent with Cloacaspis species, including Cloacaspis tesselata. The specimen shows
a minimum of 11 thoracic segments (fig. 6E).

18

AMERICAN MUSEUM NOVITATES

NO. 3928

At Little Rawhide Mountain, the “olenid” bed is at the base of the North American Whit-
erockian Stage, which puts it around the Floian-Dapingian boundary at the base of the Middle
Ordovician. The occurrences at Ike’s Canyon are not as well constrained. USGS collections
D2217-CO and D2219-CO were sampled from the Orthidiella zone (McKee et al., 1972), how¬
ever, which is correlative with Zone L of Ross (1951), the Psephosthenaspis Zone of Fortey and
Droser (1996; see also Adrain et al., 2012, for trilobite zonation) and the Rangerian zone as
defined by Ross et al. (1997). This places the Ike’s Canyon occurrences within the Dapingian.

ACKNOWLEDGMENTS

The author offers thanks to Mariah Slovacek (AMNH) for fossil preparation, and to Susan
Klofak (AMNH) and Jacob Spicer for field assistance at Ike’s Canyon and Little Rawhide Moun¬
tain in 2015. Eieldwork in Nevada was done under paleontological resource permits no. 0596-
0082 (Eorest Service) and N-93449 (Bureau of Land Management). Eurther thanks to George
Langstaff (Eorest Service) and John Kinsner (BLM) for facilitating permits, and to Kevin
(Casey) McKinney (USGS), Matt Riley (Sedgwick Museum of Earth Sciences, Cambridge),
Simon Kelly (Cambridge Arctic Shelf Programme), Eranz-Josef Lindemann (Natural History
Museum Oslo), and Jennifer Strotman (Smithsonian) for facilitating institutional visits and/or
loans of specimens. Thank you to Richard Eortey and Lisa Amati for helpful reviews. New
Spitsbergen material collected in 2016 was made possible through funding from the Niarchos
Eoundation and with the help of many people, particularly Bjorn Krdger (Einnish Museum of
Natural History), Seth Einnegan (University of California at Berkeley), Eranziska Eraneck
(NHM Oslo), and Havard Karsted (Longyearbyen); see also acknowledgements in Kroger et
al., 2017. Eieldwork in Spitsbergen was done under permit 2016/00110-2; this work is part of
Research in Svalbard (RIS) ID 10467.

REEERENCES

Adrain, J.M. 2011. Class Trilobita Walch, 1771. In Z.-Q. Zhang (editor). Animal biodiversity: an outline
of higher-level classification and survey of taxonomic richness. Zootaxa 3148: 104-109.

Adrain, J.M. 2013. A synopsis of Ordovician trilobite distribution and diversity. Memoir, Geological
Society of London 38 (1): 297-336.

Adrain, J.M., N.E.B. McAdams, and T.S. Karim. 2012. The Middle Ordovician bathyurid trilobite Pseu-
doolenoides, with a revised trilobite biostratigraphy of the Dapingian and lower Darriwilian of
western Laurentia. Zootaxa 3467: 1-97.

Brogger, W.C. 1882. Die silurischen Etagen 2 und 3 im Kristianiagebiet und auf Eker, ihre Gliederung,
Fossilien, Schichtenstorgen und Contactmetamorphosen. Kristiania [Oslo], Norway: A.W. Brogger,.

Burmeister, H. 1843. Die Organisation der Trilobiten aus ihren lebenden Verwandten entwickelt; nebst
einer systematischen Uebersicht aller zeither beschriebenen Arten. Berlin: Reimer.

Ethington, R.L., S.C. Finney, J.F. Miller, R.J. Ross, Jr., and C. Valdes-Camin. 1995. Pre-meeting trip—
central Great Basin transect. In J.D. Cooper (editor), Ordovician of the Great Basin: fieldtrip guide-

2019

HOPKINS: THE TRILOBITE SUBEAMILY BALNIBARBIINAE

19

book and volume for the seventh international symposium on the Ordovician System, Las Vegas,
NV, USA, June 1995: 1-50. Pacific Section, Society for Sedimentary Geology, Fullerton, CA.

Fortey, R.A. 1974. The Ordovician trilobites of Spitsbergen. I. Olenidae. Norsk Polarinstitutt Skrifter 160:
1-129.

Fortey, R.A. 1980. The Ordovician trilobites of Spitsbergen. III. Remaining trilobites of the Valhallfonna
Formation. Norsk Polarinstitutt Skrifter 171: 163.

Fortey, R.A., and D.L. Bruton. 1973. Cambrian-Ordovician rocks adjacent to Hinlopenstretet, North Ny
Friesland, Spitsbergen. Geological Society of America Bulletin 84: 2227-2242.

Fortey, R.A., and D.L. Bruton. 2013. Lower Ordovician trilobites of the Kirtonryggen Formation, Spits¬
bergen. Fossils and Strata 59: 1-116.

Fortey, R.A., and M.L. Droser. 1996. Trilobites at the base of the Middle Ordovician, western United
States. Journal of Paleontology 70 (1): 73-99.

Fortey, R.A., and M.L. Droser. 1999. Trilobites from the base of the type Whiterockian (Middle Ordovi¬
cian) in Nevada. Journal of Paleontology 73 (2): 182-201.

Gee, D.G., and L.M. Page. 1994. Caledoinian terrane assembly on Svalbard: new evidence from 40Ar/39Ar
dating in Ny Friesland. American Journal of Science 294: 1166-1186.

Gee, D.G., and A.M. Teben kov. 2004. Svalbard: a fragment of the Laurentian margin. Geological Society,
London, Memoirs 30 (1): 191-206.

Han, N.-R. 1983. Lower Ordovician trilobites from the Yinchufu Formation of Jiangshan, W. Zhejiang.
Acta Palaeontologica Sinica 1983-05: 571-576.

Harrison, L.B., and H.C.E. Larsson. 2015. Among-character rate variation distributions in phylogenetic
analysis of discrete morphological characters. Systematic Biology 64 (2): 307-324.

Kobayashi, T. 1936. On the Parabolinella fauna from Province Jujuy, Argentina, with a note on the
Olenidae. Japanese Journal of Geology and Geography 13 (1-2): 85-102.

Kroger, B., S. Finnegan, F. Franeck, and M.J. Hopkins. 2017. The Ordovician succession adjacent to
Hinlopenstretet, Ny Friesland, Spitsbergen. American Museum Novitates 3882: 1-28.

Lewis, P.O. 2001. A likelihood approach to estimating phylogeny from discrete morphological character
data. Systematic Biology 50 (6): 913-925.

Lisogor, K.A. 1961. [Trilobites of the Tremadoc and adjacent strata of Kendyktas]. Trudy Geologichesk-
ogo Instituta. Akademiya Nauk SSSR 18: 55-92. [in Russian]

McKee, E.H. 1976. Geology of the northern part of the Toquima Range, Lander, Eureka, and Nye Goun-
ties, Nevada. US. Geological Survey Professional Paper 931.

McKee, E.H., B.S. Norford, and R.J. Ross Jr. 1972. Gorrelation of the Ordovician shelly facies Orthidiella
zone with zones of the graptolitic facies, Toquima Range, Nevada, and North White River region,
British Golumbia. US. Geological Survey Professional Paper 800-G: G145-G156.

Monti, D.S., and V.A. Gonfalonieri. 2013. Phylogenetic analysis of the late Gambrian-early Ordovician
genus Parabolinella Brogger (Trilobita, Olenidae). Geological Journal 48 (2-3): 156-169.

Monti, D.S., and V.A. Gonfalonieri. 2017. Gomparing phylogenetics and linear morphometries to solve
the generic assignment of Parabolinella? triarthroides Harrington (Trilobita, Olenidae). Journal of
Paleontology 91 (5): 919-932.

Monti, D.S., and V.A. Gonfalonieri. 2019. First cladistic analysis of the trilobite family Olenidae from
the Furongian and Ordovician. Lethaia. [doi: 10.1111/let.l2313]

Robison, R.A., and J. Pantoja-Alor. 1968. Tremadocian trilobites from the Nochixtlan Region, Oaxaca,
Mexico. Journal of Paleontology 42 (3): 767-800.

20

AMERICAN MUSEUM NOVITATES

NO. 3928

Ronquist, R, et al. 2012. MrBayes 3.2: Efficient Bayesian phylogenetic inference and model choice across
a large model space. Systematic Biology 61 (3): 539-542.

Ross, R.J., Jr. 1951. Stratigraphy of the Garden City Formation in northeastern Utah, and its trilobite
faunas. Peabody Museum of Natural History, Yale University, Bulletin 6: 161.

Ross, R.J., Jr. 1965. Early Ordovician trilobites from the Seward Peninsula, Alaska. Journal of Paleontol¬
ogy 39 (1): 17-20.

Ross, R.J., Jr., et al. 1997. The Ibexian, lowermost series in the north American Ordovician. In M.E.
Taylor (editor). Early Paleozoic biochronology of the Great Basin, western United States. US Geo¬
logical Survey Professional Paper 1579: 1-50.

Strong, E.E., and D. Lipscomb. 1999. Character coding and inapplicable data. Cladistics 15: 363-371.

Swofford, D.L. 1998. PAUP"*^: Phylogenetic analysis using parsimony"^ 4.0. Sunderland, MA: Sinauer.

Vallance, G., and R.A. Fortey. 1968. Ordovician succession in north Spitsbergen. Proceedings of the
Geological Society of London 1648: 91-97.

Zhang, W. 1985. Synonyms of Ordovician and Cambrian trilobites. Palaeontologia Cathayana 2: 177-
178.

All issues of Novitates and Bulletin are available on the web (http://digitallibrary.
amnh.org/dspace). Order printed copies on the web from:

http;//shop.amnh.org/a701/shop-by-category/books/scientific-publications.html

or via standard mail from:

American Museum of Natural History—Scientific Publications
Central Park West at 79th Street
New York, NY 10024

© This paper meets the requirements of ANSI/NISO Z39.48-1992 (permanence of paper).