ArticlePDF Available

Placoderms (Armored Fish): Dominant Vertebrates of the Devonian Period


Abstract and Figures

Placoderms, the most diverse group of Devonian fishes, were globally distributed in all habitable freshwater and marine environments, like teleost fishes in the modern fauna. Their known evolutionary history (Early Silurian–Late Devonian) spanned at least 70 million years. Known diversity (335 genera) will increase when diverse assemblages from new areas are described. Placoderms first occur in the Early Silurian of China, but their diversity remained low until their main evolutionary radiation in the Early Devonian, after which they became the dominant vertebrates of Devonian seas. Most current placoderm data are derived from the second half of the group's evolutionary history, and recent claims that they form a paraphyletic group are based on highly derived Late Devonian forms; 16 shared derived characters are proposed here to support placoderm monophyly. Interrelationships of seven placoderm orders are unresolved because Silurian forms from China are still poorly known. The relationship of placoderms t...
Content may be subject to copyright.
EA38CH20-Young ARI 29 March 2010 14:51
Placoderms (Armored Fish):
Dominant Vertebrates of the
Devonian Period
Gavin C. Young
Research School of Earth Sciences, College of Science, The Australian National University,
Canberra ACT 0200, Australia; email:
Annu. Rev. Earth Planet. Sci. 2010. 38:523–50
The Annual Review of Earth and Planetary Sciences is
online at
This article’s doi:
Copyright c
2010 by Annual Reviews.
All rights reserved
Key Words
gnathostome origins, braincase evolution, biogeography, paleogeography,
vertebrate dispersal
Placoderms, the most diverse group of Devonian fishes, were globally dis-
tributed in all habitable freshwater and marine environments, like teleost
fishes in the modern fauna. Their known evolutionary history (Early
Silurian–Late Devonian) spanned at least 70 million years. Known diver-
sity (335 genera) will increase when diverse assemblages from new areas are
described. Placoderms first occur in the Early Silurian of China, but their
diversity remained low until their main evolutionary radiation in the Early
Devonian, after which they became the dominant vertebrates of Devonian
seas. Most current placoderm data are derived from the second half of the
group’s evolutionary history, and recent claims that they form a paraphyletic
group are based on highly derived Late Devonian forms; 16 shared derived
characters are proposed here to support placoderm monophyly. Interrela-
tionships of seven placoderm orders are unresolved because Silurian forms
from China are still poorly known. The relationship of placoderms to the two
major extant groups of jawed fishes—osteichthyans (bony fishes) and chon-
drichthyans (cartilaginous sharks, rays, and chimaeras)—remains uncertain,
but the detailed preservation of placoderm internal braincase structures pro-
vides insights into the ancestral gnathostome (jawed vertebrate) condition.
Placoderms provide the most complex morphological and biogeographic
data set for the Middle Paleozoic; marked discrepancies in stratigraphic oc-
currence between different continental regions indicate strongly endemic
faunas that were probably constrained by marine barriers until changes in
paleogeography permitted range enlargement into new areas. Placoderm
distributions in time and space indicate major faunal interchange between
Gondwana and Laurussia near the Frasnian-Famennian boundary; closure
of the Devonian equatorial ocean is a possible explanation.
Click here for quick links to
Annual Reviews content online,
• Other articles in this volume
• Top cited articles
• Top downloaded articles
• Our comprehensive search
Annu. Rev. Earth Planet. Sci. 2010.38:523-550. Downloaded from
by Australian National University on 05/30/10. For personal use only.
EA38CH20-Young ARI 29 March 2010 14:51
The Devonian Period, which lasted approximately 60 million years (420–360 Mya), witnessed
one of the most significant transformations in the history of the biosphere, at least from our hu-
man perspective as terrestrial vertebrates. The first diverse terrestrial ecosystems were already
established, and with forests evolving by the Middle Devonian, parts of our planet for the first
time would have appeared familiar to us. The first tetrapods (four-legged vertebrates), our re-
mote terrestrial ancestors, were walking around by the Middle-to-Late Devonian (perhaps much
earlier; Young 2006). The rivers, lakes, and seas teemed with aquatic jawed vertebrates (gnathos-
tomes), which had almost completely replaced the previously diverse and abundant jawless forms
(agnathans) of the Ordovician and Silurian.
Most of these aquatic vertebrates were placoderm fishes. Placoderms, with their characteristic
armor of bony plates, were the most successful and diverse group of fishes during the Devonian.
They also have an excellent fossil record because their dermal bones were generally robust and eas-
ily preserved. They included the largest Devonian vertebrates (e.g., Titanichthys from the Cleveland
Shale, probably a filter-feeder with a skull approximately 1 m long) and the smallest (Minicrania
from China, with the smallest known vertebrate head at less than 2.2 mm long; Zhu & Janvier
1996). Placoderm fossils are known from a modern latitude of nearly 80north in the Arctic
(Ellesmere Island; Kiaer 1915) and from 86south latitude in the Antarctic (Ohio Range of the
Horlick Mountains; Young 1991). They have been found at an elevation of 3800 m at Lake
Titicaca (a Late Devonian arthrodire; Diaz-Martinez et al. 1996) and from a depth of 3828 m in
the subsurface of a petroleum well in the Persian Gulf (Qataraspis; White 1969).
By the Early Devonian, placoderms were already established and diverse in both nonmarine and
marine environments on all the major continental blocks and surrounding shallow seas. A major
subgroup, the arthrodires, radiated further during the Middle and Late Devonian, particularly in
the marine environment. The arthrodires produced the largest known predators of the time, such
as Dunkleosteus and its relatives in Ohio’s Cleveland Shale and in the Upper Devonian of Morocco,
with fearsome blade-like jaws (Figures 1cand 2).
Placoderms are typically a Devonian group, and they retained their diversity until the last
Devonian stage (Famennian), after which they disappeared from the fossil record at the Devonian-
Carboniferous boundary extinction. There were no reliable pre-Devonian placoderm occurrences
when Denison (1978) wrote the last comprehensive placoderm review, but this situation has dra-
matically changed. Early Silurian placoderms from China now demonstrate a considerable evo-
lutionary history before placoderms became abundant in other geographic areas at the beginning
of the Devonian. Most of the literature on placoderms has focused on well-known groups such
as the Middle-to-Late Devonian arthrodires illustrated in Figure 1, but this represents only the
latter half of a known evolutionary history spanning 70 million years for the entire group.
The name placoderm is derived from the Greek plakos (plate) and derma (skin), referring to the
interlocking bony plates (armor) enclosing the anterior part of the body (Figure 1b). Arthrodires—
which means jointed neck, from the Greek arthron (joint) and deire (neck)—are the most diverse
subgroup, comprising 55% of named genera. Confusion has resulted from some earlier workers
using the term Arthrodira for the entire group (e.g., Stensi¨
o 1925, 1959). Under the outline clas-
sification used here (see Outline Classification of the Placodermi sidebar), the order Arthrodira is
one of nine placoderm subgroups recognized by distinctive skull patterns and other morphological
In this review I outline how our knowledge about placoderms has grown, summarize their
morphology and its significance as a model for the ancestral gnathostome condition, and briefly
discuss some of the recent controversies about their relationships and interrelationships. From a
524 Young
Annu. Rev. Earth Planet. Sci. 2010.38:523-550. Downloaded from
by Australian National University on 05/30/10. For personal use only.
EA38CH20-Young ARI 29 March 2010 14:51
Dermal skeleton:
Endoskeleton: Braincase
Skull bones Cheek Operculum Trunk-armorJaw bones
Jaw cartilage Visceral skeleton Paired ns, girdles
Unied cheek
Reduced Meckel's cartilage
Pectoral incision
Embayed trunk-armor
arch Hyoid arch
Synarcual Axial skeleton
Incision for enlarged
pectoral n
Pectoral fenestra Pelvic girdle
Pelvic girdle
Early Devonian
Middle Devonian
Late Devonian
Figure 1
Typical Arthrodira (left lateral view, not to scale) illustrating basic placoderm morphology (a–b) and interpreted major skeletal
transformations during the Devonian Period (b–d ). (a) Main components of the internal endoskeleton (based on Coccosteus). (b)The
external macromeric dermal skeleton of the same form (Eifelian Coccosteus; modified from Miles & Westoll 1968). (c) A Late Devonian
brachythoracid (based on Dunkleosteus). (d) An Early Devonian actinolepid with fin enclosed by pectoral fenestra (based on Sigaspis;
Goujet 1973). Arrows show assumed direction of evolutionary change, with associated transformations indicated, based on the
European/North American stratigraphic record, from (d) Early through (b) Middle to (c) Late Devonian. Abbreviations: ADL, anterior
dorsolateral plate of trunk-armor; MD, median dorsal plate of trunk-armor; IG, infragnathal bone; Mk, Meckel’s cartilage (primary
lower-jaw element of all jawed vertebrates); PNu, paranuchal plate of skull; SM, submarginal plate (dermal operculum). Placoderms 525
Annu. Rev. Earth Planet. Sci. 2010.38:523-550. Downloaded from
by Australian National University on 05/30/10. For personal use only.
EA38CH20-Young ARI 29 March 2010 14:51
Dermal bone
Cartilage (preserved
as surface ossication)
Figure 2
Transformation from mainly endoskeletal (a) to mainly dermal (c) structural components in the arthrodire
lower jaw (not to scale). All are lateral views of the left lower jaw (anterior to the left), except panel d,an
inner (mesial) view of the posterior end. Orange-shaded elements represent cartilage preserved as
perichondral surface ossification; the blue-shaded infragnathal (IG) is a dermal bone of the exoskeleton.
(a) Early Devonian example with complete, perichondrally ossified Meckel’s cartilage (Mk) as the main
structural component of the lower jaw (buchanosteid described by Young et al. 2001; jaw length 29 mm;
same specimen as in Figure 5e). (b) Coccosteomorph with vertically expanded dermal infragnathal bone (IG)
and cartilage unossified for most of its length, indicated by the dashed line (based on Harrytoombsia described
by Miles & Dennis 1979; jaw length 40 mm). The ossified areas at the front and back ends of Meckel’s
cartilage are called the mentomandibular (Mm) and articular (Art) bones. (c)Dunkleosteus (jaw length 26 cm),
Famennian, with cartilage reduced to a small articular ossification (Art) that carried the jaw articulation.
(d) Inner view of the posterior end of the jaw shown in panel c, showing extension of cartilage past the
posterior blade of the infragnathal. Panels cand dmodified from Heintz 1932, figures 30 and 31.
single arthrodire head, approximately 51 separate bones could finish up as rock inclusions, which
partly explains their excellent fossil record and indicates the potential complexity of placoderm
morphological data on which phylogenetic, biostratigraphic, and biogeographic hypotheses can be
established. The distribution of placoderm fishes in time and space, supported by this complex data
set, is important to our broader knowledge of the Devonian world—even if underutilized or some-
times completely overlooked. These aspects are briefly discussed in the last sections of this review.
Placoderms featured in the earliest investigations of problematic rock strata that were eventually
assigned to the Devonian System by Sedgewick & Murchison (1839). These belonged to the Old
Red Sandstone of Scotland, in which fish remains were the only conspicuous fossils. Sedgewick &
Murchison (1828) first considered placoderms to be remains of tortoises, but these were later shown
to represent two of the major placoderm subgroups: Coccosteus, an arthrodire, and Pterychthyodes
and Bothriolepis, two antiarchs. The Scot Hugh Miller (in his 1841 book The Old Red Sandstone)and
526 Young
Annu. Rev. Earth Planet. Sci. 2010.38:523-550. Downloaded from
by Australian National University on 05/30/10. For personal use only.
EA38CH20-Young ARI 29 March 2010 14:51
(modified from Denison 1978, Goujet & Young 2004)
Class Placodermi
1. Order Stensioellida
2. Order Pseudopetalichthyida
3. Order Petalichthyida
4. Order Ptyctodontida
5. Order Acanthothoraci
6. Order Rhenanida
7. Order Antiarcha
8. Order Phyllolepida
9. Order Arthrodira
Suborder Actinolepida
Suborder Phlyctaeniida
Suborder Brachythoraci
Poorly known orders, considered the most primitive by Denison (1978) but excluded from the Placodermi by other
authors (not further considered in this review).
the Estonian Herman Asmuss, who studied Middle Devonian placoderms from the Baltic region,
independently worked out that these remains (also earlier attributed to beetles) came from true
fishes. The enormous bones of the arthrodires Homostius and Heterostius, which can be scraped out
of soft, deeply weathered Devonian sandstones in the Baltic region, are still misinterpreted today
by members of the public as mammoth bones (E. Lukˇ
cs, personal communication).
Most of the placoderm orders were identified by early workers as clearly defined taxa, but
the validity of grouping them together was not apparent; some were allied with early jawless fish
(e.g., phyllolepids as heterostracans; Woodward 1915) or with subgroups of the true bony fishes
(catfish, Huxley 1861; lungfish, Woodward 1891). Agassiz (1844) synthesized existing knowledge
of fossil fishes to establish the scientific investigation of fossil early vertebrates. Specifically, four
of his genera represented four of the placoderm orders: Actinolepis, an arthrodire; Byssacanthus,an
antiarch; Chelyophorus, a ptyctodontid; and Phyllolepis, a phyllolepid.
In the century after Agassiz, the main thrust of placoderm research centered around important
fossil localities such as the Famennian Cleveland Shale (with Ohio state geologist J.S. Newberry
erecting 11 placoderm genera), the Baltic region, and Germany, the latter of which includes the
famous Lagerst¨
atten of the Hunsr¨
uckschiefer (Emsian) and Wildungen (Frasnian). From these
regions Otto Jaekel and Walter Gross together erected 28 placoderm genera ( Jaekel 1907, Gross
1932). The Swedish researcher E. Stensi¨
o published major anatomical monographs (e.g., 1963,
1969) based mainly on material from Wildungen, Podolia, and Spitsbergen. Obruchev (1964)
summarized three decades of his own research on Russian material, in addition to pioneering
work by C.H. Pander, K.E. von Eichwald, and others.
Farther afield, remains of the common Late Devonian placoderm Bothriolepis (Figure 3c) were
first recorded in Australia and Antarctica by Woodward (1916) and in China by Chi (1940). The
new method of acid extraction of bone from limestone was first documented by White (1952) on
Australian specimens from Burrinjuck, New South Wales, a locality preserving the oldest known Placoderms 527
Annu. Rev. Earth Planet. Sci. 2010.38:523-550. Downloaded from
by Australian National University on 05/30/10. For personal use only.
EA38CH20-Young ARI 29 March 2010 14:51
Ventr al
Figure 3
Restorations of various placoderms (not to scale; mostly 30–50 cm in total length). (a)Wuttagoonaspis from
central Australia (after Young & Goujet 2003). (b) Sinolepid antiarch Grenfellaspis, showing the large ventral
fenestra (opening) in the armor characteristic of Sinolepididae (Famennian, eastern Australia; from Young
1999), and the arm-like pectoral appendages characteristic of the Antiarcha. (c) Bothriolepid antiarch
Bothriolepis.(d) Whole-body preservation of the rhenanid Gemuendina from the Emsian Hunsr ¨
Germany (dorsal view; after Gross 1963). (e–f ) Whole-body preservation of the petalichthyid Lunaspis from
the Emsian Hunsr¨
uckschiefer, (e) dorsal view and (f) ventral view showing absence of pelvic fins (both after
Gross 1961). ( g) Dermal armor of the acanthothoracid Romundina, left lateral view (after Goujet & Young
2004). Abbreviations: AMD and PMD, anterior and posterior median dorsal plates of trunk-armor;
SM, submarginal plate (dermal operculum); SP, spinal plate.
528 Young
Annu. Rev. Earth Planet. Sci. 2010.38:523-550. Downloaded from
by Australian National University on 05/30/10. For personal use only.
EA38CH20-Young ARI 29 March 2010 14:51
Nu Nu
Nuchal (Nu) Central Anterior paranuchal Postmarginal (PM)
Figure 4
Skull roof dermal bone pattern in various placoderms (dorsal view, not to scale). (a) ?Phyllolepid/actinolepid
Wuttagoonaspis;(b) brachythoracid arthrodire Buchanosteus;(c) actinolepid arthrodire Actinolepis;
(d) phyllolepid Cowralepis;(e) antiarch Bothriolepis;(f) acanthothoracid Romundina;(g) petalichthyid
Lunaspis;(h) brindabellaspid Brindabellaspis;(i) quasipetalichthyid Diandongpetalichthys. Panels a,d,andg
include the dermal operculum (SM). Other abbreviations: D.end, endolymphatic duct opening; La, lateral
plate; Na, nasal opening; Nu, nuchal plate; Orb, orbit (eye socket); PM, postmarginal plate; PrM, premedian
plate. All modified or redrawn from Goujet & Young (2004), Long (1993), Ritchie (2005), Young (1979,
1980, 1988), and Zhu (1991). Placoderms 529
Annu. Rev. Earth Planet. Sci. 2010.38:523-550. Downloaded from
by Australian National University on 05/30/10. For personal use only.
EA38CH20-Young ARI 29 March 2010 14:51
clear association of fishes with a tropical reef environment. Somewhat later, Miles (1971) presented
the first placoderm monograph on the now famous Western Australian Gogo fish assemblage, also
in a Devonian reef setting (see Long & Trinajstic 2010, in this volume). Meanwhile, from China,
Liu & P’an (1958) and Liu (1963, 1973) had documented new antiarchs and petalichthyid-like
forms not readily assigned to existing taxa. In Australia, Ritchie’s (1973) first detailed study of
placoderm impressions previously assigned to the Northern Hemisphere genus Phyllolepis also
revealed a strange new skull pattern (Wuttagoonaspis;Figures 3aand 4a).
The last comprehensive review of the placoderms (Denison 1978) appeared just before some
significant papers and monographs, including Goujet (1984a) on Spitzbergen, White (1978)
and Young (1979–1981) on Burrinjuck, Zhang (1978) and Chang (1980) on China, Blieck et al.
(1980) and Janvier & Pan (1982) on the Middle East, and Goujet et al. (1984) on South America.
Placoderm research in these new regions is still in its early stages but is dramatically transforming
our understanding of evolutionary history and diversity patterns for the group. Most significant
are the new revelations from China, where nearly 60 placoderm genera—the majority endemic—
have been summarized in recent reviews (e.g., Zhu 2000, Zhu & Wang 2000, Zhao & Zhu 2007).
These extra-European areas are producing the new data that give direction to future research,
as we continue to investigate this highly diverse and interesting extinct vertebrate group, the
placoderm fishes.
The Exoskeleton versus the Endoskeleton
The fact that the vertebrate skeleton comprises two different systems (Patterson 1977) is per-
haps better displayed in the extinct placoderms than in any other gnathostome group. In contrast
to modern vertebrates, the external or dermal skeleton was clearly separated from the internal
or endoskeleton. The underlying design of a basic craniate (a chordate animal with the dorsal
nerve chord expanded at the front to form a brain) is evident in the endoskeletal components
(Figure 1a): the braincase (endocranium) housing the brain, and the vertebral column or axial
skeleton, to which the internal skeletons of the unpaired [dorsal and caudal (tail)] fins were at-
tached. Suspended beneath the braincase was a series of jointed cartilaginous arches supporting
the gills, the first (mandibular arch) much enlarged and modified to form the primary jaw cartilages
(palatoquadrate for the upper jaw, Meckel’s cartilage for the lower jaw). The second arch (hyoid
arch) was also enlarged for jaw support and to form the gill cover or operculum. These structures
represent the visceral skeleton. Encircling the axial skeleton were the two endoskeletal girdles,
which support internal skeletons of the paired pectoral and pelvic fins. These basic endoskeletal
components of an “undressed” placoderm (Figure 1a) approximate to the cartilaginous skeleton
of a modern shark.
Most placoderm orders have some early representatives that retain ossification of the en-
doskeleton, but by the Middle-to-Late Devonian internal ossification was much reduced or lost.
This loss is a serious handicap in understanding placoderm structure and function because all the
muscles were attached primitively to the endoskeleton (see below). Thus, our knowledge of the
morphology of these younger forms relies mostly on the overlying dermal skeleton.
Dermal bones can form only in the skin, and in placoderms they covered the anterior of the
body (Figure 1b), including the denticulate plates inside the mouth cavity (tooth plates of the
jaws), the sclerotic ring surrounding the outside of the eyeball, and the squamation (scale cover).
In some bones the dermal and endoskeletal constituents were closely bonded together, and a first
consideration for establishing the homology of skeletal elements is to determine the respective
530 Young
Annu. Rev. Earth Planet. Sci. 2010.38:523-550. Downloaded from
by Australian National University on 05/30/10. For personal use only.
EA38CH20-Young ARI 29 March 2010 14:51
contribution of the two discrete skeletal systems. Dermal bone can generally be identified by
its external ornament of tubercles, denticles, or ridges. All bones of the internal skeleton were
first formed as cartilage. Perichondral bone is the only form of internal ossification known in
placoderms. It is normally preserved as thin layers of smooth bone, always enclosing spaces that
in life were filled with cartilage.
The basic endoskeletal components predetermined those of the dermal skeleton, which was
developed externally as four main units (Figure 1): (a) the skull roof as a protective cover fused
to the top of the braincase, (b) the cheek unit attached to the outside of the upper jaw cartilage
(palatoquadrate), (c) the operculum in the region of the hyoid arch that covered the branchial
chamber housing the gills, and (d) the dermal shoulder girdle or trunk-armor encircling the
endoskeletal pectoral girdle (scapulocoracoid).
Sometimes the trunk-armor enclosed the pectoral fins, which emanated through a pectoral
fenestra, and in all groups the trunk-armor was connected by some form of dermal neck-joint
between the anterior dorsolateral plate (ADL) and the paranuchal plate (PNu) of the skull
(Figure 1c,d ). Many groups had enlarged spines in front of the pectoral fins (SP, Figure 3a,e),
and some had the pelvic girdle also covered by enlarged dermal bones. Enlarged plates enclosed
the pectoral fins in antiarchs to form dermally articulated appendages (Figure 3b,c), a unique
condition among vertebrates.
There is clear evidence that during the 70 million years of placoderm evolution, some en-
doskeletal structures were functionally replaced by the dermal skeleton. In the Early Devonian
Brindabellaspis, the internal pectoral girdle (scapulocoracoid) had a high scapular process (Young
1980). This was similar to the condition in modern sharks, where the process anchors the girdle
into the body musculature, as does the iliac process of the pelvic girdle as shown in Figure 1a.
However, in the most diverse placoderm group (arthrodires), the scapulocoracoid was always re-
duced to a low element inside the rigid dermal girdle, its only remaining function to provide the
articulation surface for the pectoral fins. Study of the arthrodires has documented other significant
changes in their dermal skeleton, as summarized in the next section.
Examples of Character Transformations (Arthrodira)
The best-known change in the arthrodire dermal skeleton is the reduction of the elongate, box-
like trunk-armor that enclosed the pectoral fin in typical Early Devonian arthrodires (Figure 1d).
In Coccosteus from the Eifelian (Figure 1b), representing the early Brachythoraci (meaning short
thorax), the fin was still enclosed within the pectoral fenestra, but the postpectoral wall was much
embayed (the “coccosteomorph level of organization” of Miles 1969). In advanced brachythoracids
the pectoral fenestra was replaced by a pectoral incision (Figure 1c), permitting the scapuloco-
racoid (internal shoulder girdle) to expand posteriorly, with 9–15 articulations for the basals of a
broad-based, highly mobile pectoral fin. According to Miles (1969), this “pachyosteomorph level”
was independently attained in different brachythoracid groups.
Concurrently, there was a major change in the suspension of the dermal cheek and opercu-
lum, the latter in a typical Early Devonian arthrodire being a large, ovate submarginal plate (SM,
Figure 1d) separately attached to the braincase by an underlying opercular cartilage. However,
all known brachythoracid arthrodires have the cheek and operculum combined as a single unit,
by incorporation of the submarginal plate above the dermal cheek unit (Figure 1b,c). This ef-
fectively replaced the middle connection between the upper-jaw cartilage and the braincase with
the opercular cartilage suspension. The underlying cartilaginous endoskeleton was primitively
a single, perichondrally ossified, omega-shaped palatoquadrate. Later, the nonfunctional middle
part reduced and disappeared, and the remaining anterior (autopalatine) and posterior (quadrate) Placoderms 531
Annu. Rev. Earth Planet. Sci. 2010.38:523-550. Downloaded from
by Australian National University on 05/30/10. For personal use only.
EA38CH20-Young ARI 29 March 2010 14:51
ossifications (for articulating with the braincase and supporting the jaw joint, respectively) were
held together by the rigid dermal cheek unit.
Less widely known is a similar reduction in the cartilage of the lower jaw, again involving a
functional transfer from the primary Meckel’s cartilage to the dermal infragnathal bone as the main
structural component (Figure 2). The infragnathal is robust and commonly preserved, and it has
been analyzed in several studies of large Cleveland Shale arthrodires such as Dunkleosteus,where
it is a massive bone with a prominent anterior biting portion and a strong, blade-like posterior
extension (IG, Figures 1cand 2c). However, two rarely preserved additional small bones of the
lower jaw had important functions. An internal, perichondrally ossified mentomandibular bone
at the front formed the symphysis with the jaw of the opposite side to link the left and right
branches of the jaw. The second posterior element actually carried the lower articular surface for
the jaw joint (Figure 2a,c), termed the articular bone by Jaekel (1907). In most Gogo arthrodires,
these elements are preserved as open-ended structures (Mm, Art, Figure 2b), so a cartilaginous
connection between them is assumed. A small arthrodire from Burrinjuck, New South Wales,
Australia, is the only known specimen to show the primitive condition (Young et al. 2001). In this
form Meckel’s cartilage is preserved as a single perichondral ossification (Mk, Figure 2a), which
was much larger than the slender denticulate infragnathal (IG) on its dorsal surface. There was
a large lateral depression for the adductor mandibulae muscle, but in Dunkleosteus, with Meckel’s
cartilage reduced to a small posterior knob of articular bone (Art, Figure 2c,d ), it has been assumed
that the muscle attached to the deep blade of the infragnathal bone. However, assumptions about
jaw-muscle strength based only on measurements of the infragnathal (e.g., Anderson 2008) may
be suspect because entire preservation of Meckel’s cartilage can indicate that the blade portion
was some distance in front of the jaw joint.
Friedman (2007) suggested that the placoderm infragnathal was homologous to the dentary
bone of true bony fishes (osteichthyans), but its primary position dorsal to Meckel’s cartilage
(Figure 2a) makes it positionally equivalent to the osteichthyan coronoid series [hence the term
mixicoronoid used by Stensi¨
o (1963)]. However, only a single dermal element is known in the
placoderm lower jaw, compared with numerous bones, including five coronoids, in ancestral os-
teichthyans (Zhu & Yu 2004). Therefore, homology to any one bone is difficult to sustain.
To summarize, for the known (Devonian) history of the Arthrodira, based mainly on well-
studied faunas from Europe and North America, the following four major skeletal transformations
have been generally accepted (but see below):
1. reduction of the postpectoral wall of the trunk-armor, with the pectoral fenestra enlarging to
open posteriorly as a pectoral incision, permitting a much larger, broad-based pectoral fin;
2. incorporation of the separately suspended dermal operculum into the cheek unit, so that its
internal opercular cartilage became the main articulation suspending the cheek unit from
the braincase;
3. loss of the middle portion of the omega-shaped palatoquadrate (metapterygoid region),
permitting dorsal expansion of the adductor mandibulae muscle; and
4. a much enlarged dermal infragnathal, with a vertically expanded blade portion that replaced
Meckel’s cartilage as the main structural component of the lower jaw, the cartilage being
reduced to a small posterior ossification (articular bone) that carried the jaw joint.
Non-Arthrodire Groups
For placoderm groups outside the Arthrodira, the range of distinctive body forms and skull-bone
patterns (Figures 3 and 4) gives some indication of placoderm morphological diversity. The
large submarginal plate forming the operculum in most groups (SM, Figure 3) indicates that
532 Young
Annu. Rev. Earth Planet. Sci. 2010.38:523-550. Downloaded from
by Australian National University on 05/30/10. For personal use only.
EA38CH20-Young ARI 29 March 2010 14:51
this is the more primitive of the two conditions for Arthrodira discussed above. In the skull, a
posterior unpaired nuchal plate (Figure 4) is always a major element, but groups with dorsal
nostrils (Rhenanida, Antiarcha, and some Acanthothoraci) also had a large premedian bone at the
front (PrM, Figure 4e,f ). The lateral plate of antiarchs (La) enclosed the central orbito-nasal
fenestra (an opening for both eyes and external nostrils); in some other groups, the eye sockets
formed notches in the lateral border of the skull. The Acanthothoraci and Petalichthyida had
an extra anterior pair of paranuchal plates (Figure 4f–h), whereas other groups had only one. A
primitive petalichthyid subgroup (quasipetalichthyids) is only known from South China (Figure
4i). Petalichthyids shared enclosed sensory canals and neck-joint structure with ptyctodontids.
In some acanthothoracids the sensory canals ran between the plates (Figure 4f) rather than
through the bone ossification centers, suggesting to Denison (1978) that the latter condition de-
veloped independently in placoderms and osteichthyans. The endolymphatic openings (D.end)
that connected the inner-ear cavity with the exterior were presumably primitive by outgroup com-
parison with both osteostracans (Siluro-Devonian armored agnathans) and sharks (Goujet 1984b).
However, the association in placoderms with ossification centers of large dermal skull bones is
unique—through the nuchal plate in antiarchs and Brindabellaspis and through paired paranuchal
plates in other groups. Arthrodires are characterized by an extended dermal endolymphatic canal
within the paranuchal plate. Phyllolepids (Figure 4d) had a flattened skull with a much enlarged
nuchal plate, and the remaining bones were small elements around the margins. Rhenanids and
petalichthyids were benthic forms with dorsal eyes (Figure 3d,e), the former flattened and ray-like
with enlarged pectoral fins. In contrast, all ptyctodontids had a shortened skull, enlarged eyes, and
a crushing dentition with a body form reminiscent of modern chimeras (holocephalans, a bizarre
group related to sharks).
The pelvic fins and girdle of placoderms have been a subject of recent interest, with the dis-
covery of unborn embryos inside the Gogo ptyctodontid Materpiscis by Long et al. (2008). Evi-
dence of the pelvic region is only available with exceptional whole-body preservation, as in the
uckschiefer Lagerst¨
atte where rhenanids had small semicircular pelvic fins, but the associ-
ated petalichthyid Lunaspis lacked them altogether (Figure 3d,f ). The pelvic girdle was situated
just behind the dermal trunk-armor in a typical arthrodire (Figure 1d), and in phyllolepids (Long
1984, Ritchie 2005) it was located just inside the ventral armor, with good evidence of sexual
dimorphism (Cowralepis).
The marked sexual dimorphism in the Scottish Middle Devonian ptyctodontid Rhamphodopsis
was well described by Watson (1938) and Miles (1967). Females had pelvic fins covered in large
imbricating scales, and in males the denticulate surface of the clasper elements demonstrated that
these were dermal structures. This was a key argument for concluding that sexual reproduction
developed independently in placoderms, as a specialization of one group (ptyctodontids). Chon-
drichthyans (the cartilaginous sharks and rays) had a different organization, in which the male
clasper (basipterygium) was part of the internal skeleton. However, the first acid-extracted dermal
plates of placoderm claspers, described by Miles & Young (1977), showed a strongly concave inner
face, which indicates that they must have fitted around an enlarged (unossified or nonpreserved)
basipterygium. The only difference between placoderms and chondrichthyans, therefore, was the
enlarged dermal plates—a general characteristic of the placoderm dermal skeleton (Young 1986).
On that evidence, Miles & Young (1977, p. 138) proposed a radical new interpretation: that sex-
ual reproduction in placoderms was shared with chondrichthyans and therefore primitive for the
entire group, implying that it had been secondarily lost in all non-ptyctodontid placoderms. Pos-
sible sexual dimorphism had already been suggested for two other placoderm subgroups: antiarchs
(Denison 1941) and pseudopetalichthyids (Gross 1962). Later Long (1984) proposed claspers in
the new phyllolepid Austrophyllolepis from Victoria, Australia, but the absence of any evidence from Placoderms 533
Annu. Rev. Earth Planet. Sci. 2010.38:523-550. Downloaded from
by Australian National University on 05/30/10. For personal use only.
EA38CH20-Young ARI 29 March 2010 14:51
the most diverse placoderm group (arthrodires) was the main obstacle to interpreting this mode
of reproduction for all placoderms. This is now overcome with the recent evidence of unborn
embryos in a female Gogo arthrodire (Long et al. 2009) and a single example of a preserved male
basipterygium (Ahlberg et al. 2009).
Placoderms as a Basic Gnathostome Morphotype
All modern jawed vertebrates, except the living coelacanth fish Latimeria, have a solid braincase.
In contrast, most early gnathostome groups had the braincase subdivided in the adult by cranial
fissures. These have been generally considered as specializations of the skull in different groups,
but I regard them as primitive subdivisions of a segmented structure, some of which were retained
as functional specializations (e.g., the intracranial joint of Devonian lobe-finned fishes, retained
today only in Latimeria).
By the Early Devonian, most placoderms had much of the braincase fused together as a single
block of cartilage, with little evidence of the occipital and ventral fissures that occur in basal
osteichthyans (although discrete occipital ossifications have been identified in some ptyctodontids,
petalichthyids, and possibly phyllolepids). The ventral fissure—represented in modern vertebrate
embryos as an initial subdivision between primary cartilages of the braincase floor, the trabeculae
and parachordals—may have been partly retained in some early arthrodires as a large ventral
opening named the subpituitary fenestra. However, various placoderm groups consistently show
the nasal capsules separated from the rest of the braincase by a perichondrally lined fissure that
completely subdivided the braincase through the canal for the optic nerve; this fissure is termed
the optic fissure. On the dermal skull roof, this subdivision formed a distinct transverse suture
(Figure 5e) to separate off a discrete anterior rostral capsule (often missing from preserved skulls
in primitive forms).
Separate nasal capsules may be the primitive condition for both gnathostomes and vertebrates
(in the living lamprey Petromyzon, the nasal capsules are attached to the rest of the braincase only
by connective tissue; see de Beer 1937). The placoderm condition is not known in the adults of
any living gnathostome group, and it was lost independently in several placoderm subgroups. For
example, the dermal suture for the separate rostral capsule of early arthrodires (Figure 5e) was
obliterated by extensive bone overlaps that developed progressively from the back of the skull
forward, as documented by Young (2004, figure 6).
In several Early Devonian placoderms, the internal cavities of the braincase were also
lined with perichondral bone. The long nerve canals passing through thick endocranial walls
(Figure 5a) give more detailed and robust homologies than for early osteichthyans, where cra-
nial nerves and vessels left simple openings in the walls of a high and narrow braincase. Recent
comparisons of nerve-canal length approximately 10 times longer than in Siluro-Devonian
Figure 5
Acid preparation of Early Devonian placoderms from Burrinjuck, New South Wales, Australia. (a–b) Perichondrally ossified internal
braincase cavities (buchanosteid arthrodires) in dorsal view (a, skull roof bones eroded from above) and ventral view (b, floor of
braincase broken to expose brain cavity). Cranial nerves are indicated by roman numerals; preserved structures of inner-ear cavity
include anterior and posterior ampullae (AA, PA), utriculus, and horizontal semicircular canal (HC). (c–d ) Complete left eye capsule of
the acanthothoracid Murrindalaspis in (c) outer and (d) inner views, showing foramina for all nerves and vessels servicing the eye
(described by Long & Young 1988 and Young 2008a,b). (e) Completely preserved buchanosteid head skeleton, anterior view (specimen
described by Young et al. 2001). Other abbreviations: Gr (II), groove representing one half of the canal from the optic nerve
(subdivided by the optic fissure); Hy (VII), opening for hyomandibular branch of seventh nerve (Facialis).
534 Young
Annu. Rev. Earth Planet. Sci. 2010.38:523-550. Downloaded from
by Australian National University on 05/30/10. For personal use only.
EA38CH20-Young ARI 29 March 2010 14:51
Osteostraci (fossil agnathans also with an internally ossified braincase) suggest that placoderms al-
ready had the capacity to produce the myelin sheaths required to transmit nerve impulses over such
distances (Zalc et al. 2007). The central brain cavity in an Early Devonian arthrodire (Buchanosteus)
had a relatively long occipital region (Figure 5b) but a very short forebrain (telencephalon). In the
coeval Brindabellaspis, the occipital region was short, and large paired expansions of the brain cavity
Pituitary vein
Oculomotor (III)
Hy (VII)
Gr (II)
Acousticus (VIII)
Hypophysial organ
Vagus (X)
Glossopharyngeal (IX)
Canals (spino-
occipital nerves)
Brain cavity
Optic ssure
Optic nerve (II)
Ophthalmic artery
Eye capsule
Nasal opening
Eyestalk Rostral capsule
Optic vein
Muscle scars
skull roof
Sclerotic ring
10 mm
10 mm
5 mm Placoderms 535
Annu. Rev. Earth Planet. Sci. 2010.38:523-550. Downloaded from
by Australian National University on 05/30/10. For personal use only.
EA38CH20-Young ARI 29 March 2010 14:51
Macromeric dermal skeleton
Dental lamina, true teeth
Figure 6
(a–c) The three alternative schemes of relationship of the placoderm fishes to the two major living fish
groups (osteichthyans and chondrichthyans). (d–f ) Three competing hypotheses of placoderm
interrelationships: (d) proposed by Denison (1978), (e) simplified from Goujet (1984b), and ( f) simplified
from Goujet & Young (1995, 2004).
just behind the eye sockets (Young 1980) housed either optic lobes of the mesencephalon or the
cerebellum (metencephalon). The structural similarity to osteostracan and galeaspid agnathans
suggests that this was an inherited ancestral condition for gnathostomes (Forey & Janvier 1994,
figure 11; Janvier 1996, p. 274).
Detailed comparative studies by the Swedish researcher Erik Stensi ¨
o (e.g., 1925, 1963) estab-
lished basically the same morphological relations of cranial nerves and vessels as worked out for
living fishes from comparative anatomy and embryology (e.g., Goodrich 1930, de Beer 1937).
o interpreted some features (eyestalk, external course of the orbital artery, internal connec-
tion between the efferent pseudobranchial and internal carotid arteries) to indicate chondrichthyan
affinity for placoderms (see Figure 6a). However, subsequent study of acid-prepared specimens
(Young 1979, 1980, 1986) demonstrated either that placoderms resembled osteichthyans rather
than chondrichthyans or that they displayed the primitive gnathostome condition.
The orbital cavity (eye socket) in several placoderm groups shows a central unossified area
of varying shape behind and below the optic nerve foramen, the attachment site for a carti-
laginous eyestalk as in modern sharks and rays, but now also demonstrated in several early os-
teichthyans (Basden et al. 2000). Depressions in the wall of the eye socket named myodomes
(sites of muscle attachment) give reliable evidence for the arrangement and innervation of the
536 Young
Annu. Rev. Earth Planet. Sci. 2010.38:523-550. Downloaded from
by Australian National University on 05/30/10. For personal use only.
EA38CH20-Young ARI 29 March 2010 14:51
extrinsic eye muscles. In some forms a perichondrally ossified sclerotic cartilage encapsulated the
eye (Figure 5c,d ), preserving the attachment for the other end of the eyestalk, muscle scars for the
extrinsic eye muscles, and openings for all nerves and vessels that passed into the eyeball. These
data demonstrate a nerve and muscle arrangement in placoderms that differs from that of all other
gnathostomes, but in some aspects resembles that of the living jawless lamprey (Young 2008a,b).
To view videos of the specimens illustrated in Figure 5b–d, follow the Supplemental Materials
link from the Annual Reviews home page at
The few examples just outlined are presented here to give an indication of the amazing mor-
phological detail that can be extracted from exceptionally preserved extinct fossil vertebrates such
as the placoderms. Placoderms (and some other early vertebrate groups) make exquisite exem-
plars of evolutionary processes that operated hundreds of millions of years in the past, the same
processes that produced the enormous diversity of living vertebrates. However, such detailed evi-
dence provided by the fossil record has been rather underutilized (see Janvier 1996, pp. 285–287),
particularly in recent years as new tools such as molecular and developmental biology have been
applied to the investigation of evolutionary history. For example, molecular systematics of extant
taxa has been advocated as the new approach to understand ancient “rapid radiations” (Whitfield
& Lockhart 2007). How such techniques would decipher the rapid gnathostome diversification of
the Early-to-Middle Paleozoic using only molecular data from extant taxa, without taking into ac-
count the evidence provided by highly diverse but entirely extinct groups such as the placoderms,
is not at all clear.
Applying the principles of classical comparative anatomy, Stensi¨
o (1931, 1934) demonstrated that
two groups (antiarchs and phyllolepids), previously regarded by many workers as agnathans, were
in fact placoderms and thus true jawed vertebrates. Placoderm monophyly became the established
view during the following decades, but a few years ago some of the old ideas were revived for one
of these groups, the antiarchs.
Johanson (2002) suggested that antiarchs shared primitive features of the pectoral girdle with
jawless osteostracans, and that non-antiarch placoderms could therefore be more closely related to
other jawed fishes (that is, that the placoderms are paraphyletic). Johanson & Smith (2005, p. 311)
raised the further possibility that some arthrodires had true teeth and thus could be more closely
related to crown-group gnathostomes than other placoderms, thereby questioning monophyly
both of the Placodermi and of the Arthrodira. Recently a 194-character parsimony analysis of
early osteichthyan interrelationships by Friedman (2007, figures 11 and 12) has incorporated
these ideas to place the antiarch Bothriolepis as immediately related to osteostracan agnathans,
implying that placoderms are no more than a paraphyletic assemblage of basal jawed vertebrates
(Friedman 2007, p. 310). Similarly, an analysis by Brazeau (2009, figure 3) has placed antiarchs
as a sister group to a clade comprising other placoderms and other gnathostomes, also indicating
placoderm paraphyly.
A major difficulty with these reinterpretations is that they are based on analysis of a few highly
derived taxa from the Late Devonian. The evidence of placoderm “teeth” (with a pulp cavity)
comes from a single Gogo arthrodire, whereas detailed investigations by many earlier researchers
(already summarized by Heintz 1932) concluded that the cusps and denticles on arthrodire gnathal
elements were completely different from the teeth of osteichthyans. Shedding of denticles on
placoderm gnathal elements has never been observed, indicating that a dental lamina had not
developed (Reif 1982, Janvier 1996). Numerous earlier studies (reviewed by Young 2003a, 2009)
reached the conclusion that placoderms were extinct jawed vertebrates that lacked true teeth, thus Placoderms 537
Supplemental Material
Annu. Rev. Earth Planet. Sci. 2010.38:523-550. Downloaded from
by Australian National University on 05/30/10. For personal use only.
EA38CH20-Young ARI 29 March 2010 14:51
demonstrating that “dentitions evolved after the evolution of jaws” (Reif 2002, p. 57). Studies
of Early Devonian arthrodires from Australia (Young et al. 2001; our unpublished CT scanning
studies with T. Senden on exceptional new examples) overwhelmingly support that view as a
refutation of the alternative teeth-before-jaws hypothesis that was derived from ideas about the
vertebrate affinities of conodonts.
Similarly, reassessment of the antiarchs has been based primarily on Bothriolepis canadensis,
a Frasnian species that existed 50 million years after the first appearance of antiarchs in the
fossil record (Early Silurian). Morphological comparisons that include primitive Early Devonian
antiarchs (Young 2008c) reveal spurious homologies, demonstrating that Bothriolepis canadensis did
not share with osteostracans a “pectoral glenoid situated anterior to the posterior wall of the gill
chamber,” as used by Brazeau (2009, character 122). The special placoderm tissue semidentine is
also recorded in Early Devonian antiarchs (Young 2008c; cf. Brazeau 2009, character 5), although
it is lost in most or all younger members of the group, including Bothriolepis canadensis.Toanalyze
the cellular dermal bone tissue of this highly derived, stratigraphically late placoderm taxon as an
exemplar of the primitive gnathostome condition, as proposed by Downs & Donoghue (2009),
thus lacks any justification.
Another key character used by both Friedman (2007, character 190) and Brazeau (2009, char-
acter 117) is the assumed absence of pelvic fins in Bothriolepis canadensis, interpreted as a primitive
feature shared with osteostracan agnathans. This is a major reinterpretation from Stensi¨
o’s (1948,
figures 59 and 60) earlier comparative argument that a pelvic fin must have been present, even if
secondarily reduced. New evidence (e.g., Parayunnanolepis from China; Zhang et al. 2001) now
confirms the absence of pelvic fins in Early Devonian antiarchs, but numerous other antiarch
features, by which Stensi¨
o (1931) demonstrated their placoderm affinities, stand against the inter-
pretation that this absence is primitive. The pelvic fin and girdle are well known in other groups
such as arthrodires, ptyctodontids, and acanthothoracids, so the lack of pelvic fins in some (e.g.,
the petalichthyid Lunaspis from the Hunsr¨
uckschiefer; Figure 3f) can only be interpreted as a
secondary loss.
To conclude this section, I present an updated list of 16 synapomorphies (shared derived
characters) that define the Placodermi as a monophyletic group (see Synapomorphies sidebar).
Only one of these does not apply to the order Antiarcha. It is noted, however, that polarity for
some characters in the list will depend on how the placoderms are assumed to be related to other
fishes (discussed below).
Molecular studies offer significant new data for phylogenetic analysis, but molecular data will
probably never be available for entirely extinct groups such as placoderms. Nevertheless, molecular
research can influence our understanding of how fossil groups are related to living taxa.
Living jawed fishes belong in two major groups—osteichthyans (bony fishes) and chon-
drichthyans (cartilaginous fishes). Each has a distinct fossil record extending back into the Early
Paleozoic, which completely contradicts recent molecular studies that place chondrichthyans as a
subgroup of the bony fishes (e.g., Arnason et al. 2004; cf. Hallstr ¨
om & Janke 2009).
Osteichthyans are characterized by their dermal skeleton of large overlapping or interlocking
bones that form a consistent pattern, a condition termed macromeric. This contrasts with the
micromeric dermal skeleton of living and fossil chondrichthyans, whose skin is embedded with
numerous small tooth-like or scale-like elements (odontodes of Reif 1982). The possession of bony
armor in placoderms establishes a prima facie case for allying them with the living bony fishes,
but this would require independent evolution of true teeth and a dental lamina in osteichthyans
538 Young
Annu. Rev. Earth Planet. Sci. 2010.38:523-550. Downloaded from
by Australian National University on 05/30/10. For personal use only.
EA38CH20-Young ARI 29 March 2010 14:51
(modified from Goujet 2001; Goujet & Young 2004; Young 2008a,c, 2009)
1. Distinctive pattern of dermal bones in skull roof, cheek, and operculum
2. Paired external openings of endolymphatic ducts linked to dermal bone ossification centers
(nuchal or paranuchal plates) in posterior part of the skull roof
3. Simple jaws with only two or three pairs of bony tooth plates
4. Omega-shaped upper-jaw cartilage (palatoquadrate), with deep ventral embayment for ad-
ductor mandibulae muscle (which was not enclosed mesially by dermal bone; see 6, 8)
5. Palatoquadrate enclosed by lateral attachment to one or two dermal bones (suborbital,
postsuborbital plates), neither carrying socketed teeth (cf. maxilla, quadratojugal of oste-
6. Mesial surface of palatoquadrate lacking dermal bone cover (cf. entopterygoid of oste-
7. Palatoquadrate carrying single dermal gnathal element (posterior supragnathal) (cf. der-
mopalatine, ectopterygoid of osteichthyans, both of which carry true teeth)
8. No dermally enclosed adductor fossa in upper jaw
9. Lower-jaw cartilage (Meckel’s cartilage) carrying one dermal bone (infragnathal) with a
primitive position on its dorsal face (cf. multiple coronoids of osteichthyans) and lacking
external and internal cover of dermal bones (cf. dentary, prearticular of osteichthyans)
10. Adductor mandibulae muscle with broad lateral attachment to Meckel’s cartilage (cf. poste-
riorly confined adductor fossa of osteichthyans)
11. A special type of opercular suspension, comprising a dermal submarginal plate connected
directly to the braincase via a cartilage of presumed hyoid-arch derivation
12. Extensive postorbital endocranial processes fused to the inner side of the dermal skull roof
to delineate muscle attachments for operculum, visceral arches, and shoulder girdle
13. Branchial chamber confined beneath braincase by anteroventrally sloping dermal post-
branchial lamina
14. Exoskeletal shoulder girdle including one or two median dorsal elements overlapped with
interlocking lateral and ventral plates to form a rigid ring encircling the trunk
15. Dermal articulation between skull and shoulder girdle localized to paired dermal neck-joint
between anterior dorsolateral and paranuchal plates
16. Special hard tissue (semidentine) in surface layer of dermal elements
and chondrichthyans (Figure 6c). Most recent analyses (Young 1986, Goujet & Young 1995,
Goujet 2001, Goujet & Young 2004, Friedman 2007, Young 2008c) maintain the alternative view
that placoderms are more likely the sister group to the remaining gnathostomes (Figure 6b). This
implies that the macromeric dermal pattern developed independently from a micromeric common
ancestor in both placoderms and osteichthyans. Extensive skull-bone overlaps as in osteichthyans
evidently developed within the arthrodires, and general differences in the various skull patterns of
placoderms compared with those of osteichthyans (Figure 4) include the unpaired nuchal plate
in the posterior skull margin (normally paired bones in this position in osteichthyans), the paired
endolymphatic duct openings, and fewer bones around the orbits and in the cheek and gill cover.
In addition, the dermal shoulder girdle of typical osteichthyans is more flexibly attached to the Placoderms 539
Annu. Rev. Earth Planet. Sci. 2010.38:523-550. Downloaded from
by Australian National University on 05/30/10. For personal use only.
EA38CH20-Young ARI 29 March 2010 14:51
back of the skull, and it lacks the median dorsal plate that in typical placoderms forms a rigid ring
encircling the trunk.
However, three recent discoveries from the Siluro-Devonian of China now provide contrary
evidence to the view that a “fundamental difference in pattern” (Young 1986, p. 7) demonstrated
that large dermal bones had evolved independently in placoderms and osteichthyans. Zhu et al.
(1999) and Zhu & Schultze (2001) discussed the basal osteichthyan Psarolepis, the first to show
separate spinal elements in front of the pectoral fins, as in placoderms. They suggested that the ex-
tracleithrum of primitive coelacanths might also be homologous, implying at least that the dermal
shoulder girdle of osteichthyans and placoderms could have evolved from a macromeric ancestral
condition (as in Figure 6c). Zhu et al. (2006) described another basal osteichthyan, Meemannia,
which apparently combines characters of the two major osteichthyan subgroups (Actinopterygii
and Sarcopterygii) but also resembles the placoderms, with its rostral region separated from the
rest of the skull and braincase (also seen in the early osteichthyan Dialipina; Schultze & Cumbaa
2001, figure 18.2). Recently Zhu et al. (2009) described the oldest known articulated bony fish,
Guiyu from the Late Silurian of China. In its skull and cheek pattern and its possession of dentiger-
ous jaw bones, Guiyu is typically osteichthyan. However, its dermal shoulder girdle includes both
a prepectoral spinal plate (as in Psarolepis) and two large median dorsal plates behind the back of
the skull. The median bones have internal keel-like structures, like the median dorsal plate of pla-
coderms, and the posterior element is larger and carries a spine, as in the earliest antiarchs and in
the oldest known placoderms from the Chinese Silurian (Zhu & Wang 2000). The single median
dorsal plate of most placoderms was previously considered to indicate the primitive placoderm
condition, but an additional posterior median element (typical of all antiarchs) is also known in
some acanthothoracids (Figure 3g) and ptyctodontids (Long 1997). The new evidence of Guiyu
as an outgroup to placoderms now strongly indicates that the possession of two median dorsal
elements was primitive for placoderms and that this pattern could have originated in a macromeric
common ancestor of placoderms and osteichthyans. This would imply either secondary loss of the
macromeric dermal skeleton in chondrichthyans, or independent evolution of teeth and a dental
lamina (Figure 6c). Resolution of these major questions must await detailed documentation of the
Silurian forms from China.
Current competing hypotheses (Figure 6d–f ) depend on how the two most diverse placoderm
orders (arthrodires and antiarchs) are related. That this major question remains unresolved results
from the fact that most of the data represent only the latter half of placoderm evolutionary history.
Description of Chinese Silurian placoderms will have a major impact on character analysis, but
some underlying assumptions of the existing hypotheses can be outlined. Two of the four schemes
compared by Young (1986, figure 18), with ptyctodontids in a basal position, no longer apply be-
cause we now know that internal fertilization was more widespread in placoderms than previously
thought (see above). Denison (1978) considered the elongate, box-like trunk-armor a shared
derived character of arthrodires and antiarchs (Figure 6d). The enclosed pectoral fin as another
synapomorphy is supported by the small, separate anterior lateral plate in the Early Devonian
antiarch Phymolepis, indicating the antiarch pectoral attachment is equivalent to the arthrodire
pectoral fenestra (Young & Zhang 1992, 1996). Both groups also have a somewhat hexagonal
skull shape, with lateral corners formed by large postmarginal (PM) plates (color-coded in
Figure 4b,c,e). This contrasts with the parallel-sided skull of other groups (Figure 4f,h,i ).
Goujet’s (1984b) alternative proposal (Figure 6e) identified dorsal nasal openings as a shared
derived character uniting rhenanids, antiarchs, and some acanthothoracids (which are probably
540 Young
Annu. Rev. Earth Planet. Sci. 2010.38:523-550. Downloaded from
by Australian National University on 05/30/10. For personal use only.
EA38CH20-Young ARI 29 March 2010 14:51
paraphyletic; Goujet & Young 2004). The “acanthothoracid” Brindabellaspis is an intermediate,
with nasal openings in the anterodorsal corner of the eye sockets, but is also the only known
placoderm (apart from all antiarchs) with paired endolymphatic openings at the nuchal ossification
center (Figure 4h). As in most other placoderm groups, the acanthothoracid trunk-armor was
short and the pectoral fin was not enclosed (Figure 3g); this was assessed by Denison (1978) as
primitive by outgroup comparison with other gnathostomes. However, with placoderms placed
as the gnathostome sister group (Figure 6b), the appropriate outgroup becomes osteostracan
agnathans, and this argument would not apply. An assumption that the dorsal nasal openings
of osteostracans was also the primitive condition for placoderms (Goujet & Young 2004)
would give the arrangement of Figure 6f. However, this carries the implication that the lateral
nasal openings of other placoderms, also the standard arrangement for both osteichthyans and
chondrichthyans, would need to be independently evolved. Thus there is no clear resolution of
the interrelationships of major placoderm orders, perhaps because the evidence is derived largely
from the latter part of placoderm evolutionary history. It is hoped that documentation of crucial
new forms from the Siluro-Devonian of China will provide some clarification in the future.
A final important point concerns the uncertainties discussed in the previous section about how
the placoderms may be related to the other major fish groups and the impact this has on assump-
tions of character polarity within the placoderms. A key example is the assumed character trans-
formation within arthrodires from long to short trunk-armor, as outlined above and illustrated in
Figure 1. This orthodox view was not accepted by the great Swedish researcher Stensi ¨
o (1959,
1963, 1969), whose “purely anatomical approach” concluded that the free pectoral fin of Late
Devonian brachythoracids was primitive rather than derived because it resembled that of the
osteichthyan outgroup. Miles (1971, p. 227) rejected Stensi¨
o’s arthrodire phylogeny on the
grounds that it required a fossil record “so strongly biased that the true evolutionary sequences
are ...actually reversed.” However, such a bias has been demonstrated for the antiarchs, which
underwent a dramatic radiation in the Silurian and Early Devonian of China, before the group
even appeared in European Devonian sequences (see discussion below).
Could such a stratigraphic bias also apply to the arthrodires? “Advanced” brachythoracids (with
a pectoral incision) were already known to occur rarely in the Early Devonian (Miles 1969). Since
then, the poorly known, oldest placoderms from the Silurian of China have been discovered and
are now recorded to include undescribed arthrodire remains (Zhu & Wang 2000), so this must
now be considered an open question. Until the much older antiarch and arthrodire-like forms
from China are better documented, little progress can be made on resolving the conflicting data
sets obtained from the more derived Devonian placoderm groups.
Currently there are approximately 335 validly named placoderm genera, some represented by
complete skeletons but most based on just the large or more robust bones, so a sampling bias
against small or fragile skeletons seems unavoidable. Except for a few (Bothriolepis,Asterolepis,
Coccosteus), most genera are monospecific. Of the approximately 49 species assigned to Coccosteus,
Denison (1978) regarded 7 at most to be valid, but it is likely that a reasonable proportion of the
60 species of Bothriolepis are valid (given its wide distribution and Emsian-Famennian stratigraphic
Some subjectivity in dealing with fossil species seems unavoidable, and researchers approach
the issue in two different ways, some tending to define more species (the splitters) than oth-
ers do (the lumpers). Exceptional preservation of many complete specimens at some localities
(e.g., the Australian phyllolepid Cowralepis described by Ritchie 2005 and the Gogo arthrodires Placoderms 541
Annu. Rev. Earth Planet. Sci. 2010.38:523-550. Downloaded from
by Australian National University on 05/30/10. For personal use only.
EA38CH20-Young ARI 29 March 2010 14:51
described by Trinajstic & Dennis-Bryan 2009) can accentuate the problem of distinguishing intra-
from interspecific variation. Many well-preserved specimens may demonstrate gradations between
osteological variations that have elsewhere been considered diagnostic for incomplete fossil taxa.
However, that only one species is represented in such samples is an underlying assumption that can
influence the result. Most modern fish species are immediately recognizable by differences in sur-
face color or pattern, information not available in fossils. Even the chondrichthyans, whose species
are generally drab, have numerous examples of similarly shaped species with distinctive markings
(e.g., many rays; species of the wobbegong shark Orectolobus). These would never be distinguished
as fossils, even with whole-body preservation such as that occurring in the Hunsr ¨
atte. For this reason, I consider fish diversity at the species level in the fossil record prob-
ably to be greatly underestimated.
Taking this into account, and recognizing the reduced likelihood of preservation for many
small species with fragile skeletons, I compared placoderm species, genera, and families in an
Early Devonian tropical reef environment (Burrinjuck, New South Wales) with teleosts on the
modern Great Barrier Reef (Young 2009), and found no compelling evidence of significantly less
Currently unexplained is the higher diversity of some Devonian fish assemblages in geographic
regions where fish faunas have been more recently described, such as Australia and China, com-
pared with the classic localities of Europe and North America. For example, the Scottish Middle
Devonian Old Red Sandstone has a fauna of 19 genera, of which just less than half are placo-
derms (Newman & Trewin 2008). In the Frasnian Lagerst¨
atte of Miguasha (Schultze & Cloutier
1996) there are 21 fish genera, but only two placoderms. In contrast, the Burrinjuck tropical reef
assemblage (Emsian) mentioned above comprises nearly 50 genera recognized so far, of which
80% are placoderms (Young 2009). The Aztec fish fauna of Antarctica has 33 genera and 45
species (Young & Long 2005, table 1); 47% of these species are placoderms. Although Denison
(1978) listed only 17 genera of antiarchs from throughout the world, Zhu (2000) listed 23 an-
tiarchs just from the Devonian of Asia, and the phylogenetic analysis by Zhu (1996) dealt with
46 antiarch genera and 154 species from all known localities. Placoderm specialists are also aware
of diverse undescribed assemblages from regions such as the Middle East and central Asia (e.g.,
evre et al. 1993), or Gondwana (e.g., Young et al. 2010). Thus, a significant increase in generic
diversity seems a certain outcome of future systematic descriptions. Clearly, regarding placoderms
as the “teleosts of the Devonian” seems an apt comparison ( Janvier 2007, p. 33) if one takes into
account that diverse endemic placoderm assemblages from regions outside Europe and North
America still await documentation (see next section).
Early vertebrate research originated within the scientific tradition of Europe and North America.
As Devonian rocks were explored in more remote parts, the ground plan seemed secure, such that
a “law of diminishing returns in exploration” (Westoll 1958) was reiterated by Miles (1971, p. 227)
in his first monograph on the Gogo placoderms: “new material tends more and more to confirm
what has already been known or suspected.”
However, incipient cracks in these foundations had already been overlooked. Wang (1943)
noted that the Late Devonian age assigned to the antiarch Bothriolepis from China, based on com-
parisons with Europe, was contradicted by interbedded invertebrate assemblages. Skepticism of
such an age disparity persisted for decades (e.g., Westoll 1979), but modern evidence demonstrates
a first appearance for Bothriolepis in the Emsian of China (Zhu et al. 2000, figure 2), compared
with a late Givetian first appearance in the Baltic region of the Old Red Sandstone paleocontinent
542 Young
Annu. Rev. Earth Planet. Sci. 2010.38:523-550. Downloaded from
by Australian National University on 05/30/10. For personal use only.
EA38CH20-Young ARI 29 March 2010 14:51
(e.g., Mark-Kurik 2000). This is just one of many examples of temporal discordance in the placo-
derm stratigraphic record summarized in Figure 7. A different scenario is indicated by another
major antiarch subgroup, the Sinolepididae, first recognized when the distinctive Late Devonian
form Sinolepis was erected by Liu & P’an (1958). The group was later shown to extend back to the
Early Devonian with several older genera in South China and northern Vietnam (e.g., Xichonolepis,
Dayaoshania,Vanchienolepis). One sinolepid genus, Grenfellaspis (Figure 3b), is known only from
the latest Devonian (Famennian) of East Gondwana, but sinolepids have never been recorded in
older Gondwana assemblages nor in any Devonian sequence of Laurussia.
Another antiarch example is the genus Remigolepis, first described (Stensi ¨
o 1931) from the late
Famennian of East Greenland (the famous Ichthyostega-Acanthostega tetrapod locality). Since then,
the only other confirmed occurrence in Laurussia (Lukˇ
cs 1991) is another Famennian locality
also yielding tetrapods—the genus Tulerpeton from Tula, Russia. On the other hand, Remigolepis
is diverse in strata at least as old as Frasnian in China (Pan et al. 1987) and Australia (Young 1999,
2007), in both areas again associated with Devonian tetrapods (Metaxygnathus, Campbell & Bell
1977; Sinostega, Zhu et al. 2002).
The order Phyllolepida demonstrates the greatest time/space disjunction for Devonian fishes.
In 165 years of research since Agassiz (1844) erected the genus Phyllolepis, only nine Famennian
species in this genus have been recognized in Laurussian sequences (again including the Greenland
tetrapod localities). In contrast, 25 years of research in East Gondwana (Long 1984; Ritchie 1984,
2005; Young 2005a,b,c; Young & Long 2005) has documented five phyllolepid genera so far
(Austrophyllolepis, Cobandrahlepis, Cowralepis, Placolepis, Yurammia), all pre-Famennian. The oldest
phyllolepid record (Emsian of Saudi Arabia; Leli`
evre et al. 1999) and the similar concentric-ridged
ornament shared with the East Gondwana endemic Wuttagoonaspis point to a Gondwana origin
and subsequent dispersal into the Northern Hemisphere (Young 2005a). Although Dupret (2008)
and Dupret & Zhu (2008) report putative early representatives of these groups from China based
on single incomplete skulls, phyllolepids and wuttagoonaspids are absent from well-documented
younger Chinese sequences, so more evidence is needed to support these claims.
These disjunctions indicate complex dispersal patterns between different regions during the
Devonian (Figure 7). Obviously, purely stratigraphic global databases used to determine ex-
tinction rates (e.g., Sepkoski 2002) cannot accommodate the complexities just summarized for
different regions. Generic diversity for one placoderm subgroup (antiarchs) in China (Zhao &
Zhu 2007) revealed first a low-diversity origination interval for most of the Silurian, then a ma-
jor radiation in the Early Devonian, succeeded by two alternating survival and recovery intervals
(Middle-to-Late Devonian). At the same time, phyllolepids in Gondwana were experiencing a
Givetian-Frasnian radiation (Figure 7). Globally, placoderms have been cited for high extinc-
tion rates in the Givetian-Frasnian (e.g., more than 65% for the late Frasnian; Bambach 2006,
p. 137), but supporting data are incomplete (only 164 placoderm genera are in the online Sep-
koski marine genera data set, and key antiarch taxa—Asterolepis,Bothriolepis, and Remigolepis—are
Biogeography must be considered in the analysis of global biostratigraphic data for highly
mobile animals such as vertebrates. Rather than demonstrating significant extinction during the
major Frasnian-Famennian Kellwasser event, the placoderm evidence indicates a biotic dispersal
episode between southern and northern land masses, also affecting other fish groups (Janvier
ement 2005) and possibly involving tetrapods, for which associated placoderms could be a
proxy (Young 2006). Major paleogeographic rearrangement has been proposed to explain this,
involving Late Devonian closure of the equatorial ocean that earlier separated Gondwana from
Laurussia. This would imply dramatically modified oceanic-circulation patterns, perhaps driving
global climatic and atmospheric change as a backdrop to full terrestrialization of the biota. Research Placoderms 543
Annu. Rev. Earth Planet. Sci. 2010.38:523-550. Downloaded from
by Australian National University on 05/30/10. For personal use only.
EA38CH20-Young ARI 29 March 2010 14:51
Placodermi (undescribed)
Late Devonian
3-5 MAV 6-9 MAV13
MAV 0-2
Middle Late
Millions of years ago
China East Gondwana Laurussia Global
544 Young
Annu. Rev. Earth Planet. Sci. 2010.38:523-550. Downloaded from
by Australian National University on 05/30/10. For personal use only.
EA38CH20-Young ARI 29 March 2010 14:51
continues to truly globalize the complex placoderm data set with which to test and develop these
ideas, as summarized below.
1. Elucidation of the structure of the earliest known placoderms from the Silurian of Asia.
2. Systematic documentation of large undescribed placoderm assemblages (central Asia,
Middle East, Australia, and other poorly studied regions).
3. A better supported scheme of placoderm interrelationships based on these data.
4. Application of a robust placoderm phylogeny to more detailed analysis of placoderm
distribution in space and time during their known 70-Ma fossil record.
5. Application of this highly complex data set to a better understanding of changes in global
paleogeography and climate that characterized a crucial period in Earth history, the
complete terrestrialization of the biota during the Devonian Period.
The author is not aware of any affiliations, memberships, funding, or financial holdings that might
be perceived as affecting the objectivity of this review.
I thank fellow early vertebrate researchers, particularly placoderm specialists, for numerous dis-
cussions and free access to unpublished information for many years. They include R. Carr, Chang
M.M., D. Goujet, A. Ivanov, P. Janvier, H. Leli`
evre, Liu Y.H., J. Long, E. Lukˇ
cs, R.S. Miles,
Barwick gave assistance and advice on graphics. Z.X. Li made available Devonian paleomagnetic
reconstructions. Research has been supported by Australian Research Council Discovery Grants
DP0558499 and DP0772138. Dedicated to the memory of Wolf-Ernst Reif, who died during
preparation of this paper, on June 11, 2009.
Figure 7
Disparate Devonian stratigraphic ranges for placoderms between Asia, Gondwana, and Laurussia, and postulated biotic-dispersal/
range-enlargement episodes (A–D in square boxes) to explain them. (a) Stratigraphic subdivision and numerical calibration (left side)from
ICS Stratigraphic Chart (2008). Silurian (1–7) and Devonian (I–XI ) Chinese macrovertebrate assemblages after Zhu & Wang (2000)
and Zhu et al. (2000). East Gondwana macrovertebrate (MAV) zonation shown for the Devonian (from Young 2003b figure 5). Width
of stratigraphic bars approximately indicates diversity (for more detail on China, see Zhu 2000 figure 3). Range-enlargement events are:
S-D, Siluro-Devonian boundary global expansion of arthrodires and ?petalichthyids. A, Pragian-Emsian exchange between Asia and
East Gondwana (A1), including asterolepid antiarchs that expanded in the Eifelian to Laurussia (A2). B, range enlargement of
Bothriolepis from its Emsian earliest occurrence in South China (B1, extension into East Gondwana; B2, extension into Laurussia).
C, Famennian placoderm expansion into Laurussia indicated by antiarchs (Remigolepis) and phyllolepids (Phyllolepis); other groups
possibly associated are groenlandaspids, rhizodontids, gyracanthids, and tetrapods. D, late Famennian expansion of sinolepid antiarchs
from China into East Gondwana. (b) Generalized placoderm localities on a paleomagnetic reconstruction provided by Z.X. Li (updated
from Young 2003b figures 1 and 3, which give locality details). Devonian dispersal episodes (probably not unidirectional) correspond to
those of panel a. Locality symbols for key taxa (Bothriolepis, phyllolepids, sinolepids) are filled for older occurrences, and open for
younger (postdispersal) occurrences. Abbreviations: KAZ, Kazakhstan; NC, North China block; SC, South China block. Placoderms 545
Annu. Rev. Earth Planet. Sci. 2010.38:523-550. Downloaded from
by Australian National University on 05/30/10. For personal use only.
EA38CH20-Young ARI 29 March 2010 14:51
Agassiz L. 1844. Monographie des poissons fossiles du vieux gr`es rouge du syst`eme D´evonien (Old Red Sandstone) des
Iles Britanniques et de Russie.Neuch
atel: Jent & Gassmann. 171 pp.
Ahlberg PE, Trinajstic K, Johanson Z, Long JA. 2009. Pelvic claspers confirm chondrichthyan-like internal
fertilization in arthrodires. Nature 460:888–89
Anderson P. 2008. Shape variation between arthrodire morphotypes indicates possible feeding niches.
J. Vertebr. Paleontol. 28:961–69
Arnason U, Gullberg A, Janke A, Joss J, Elmerot C. 2004. Mitogenomic analyses of deep gnathostome diver-
gences:afishisafish.Gene 333:61–70
Bambach RK. 2006. Phanerozoic biodiversity mass extinctions. Annu. Rev. Earth Planet. Sci. 34:127–55
Basden A, Young GC, Coates M, Ritchie A. 2000. The most primitive osteichthyan braincase? Nature 403:186–
Blieck A, Golshani F, Goujet D, Hamdi B, Janvier P, et al. 1980. A new vertebrate locality in the Eifelian of
the Khush-Yeilagh Formation, eastern Alborz, Iran. Palaeovertebrata 9:133–54
Brazeau M. 2009. The braincase and jaws of a Devonian ‘acanthodian’ and modern gnathostome origins.
Nature 457:305–8
Campbell KSW, Bell MW. 1977. A primitive amphibian from the Late Devonian of New South Wales.
Alcheringa 1:369–81
Chang MM. 1980. Preliminary note on a Lower Devonian antiarch from Yunnan, China. Vertebr. PalAsiat.
18:179–90 (In Chinese and English)
Chi YS. 1940. On the discovery of Bothriolepis in the Devonian of Yunnan. Bull. Geol. Soc. China 20:57–72
de Beer GR. 1937. The Development of the Vertebrate Skull. Oxford: Oxford Univ. Press. 554 pp.
Denison RH. 1941. The soft anatomy of Bothriolepis.J. Paleontol. 15:553–61
Denison RH. 1978. Handbook of Paleoichthyology, ed. H-P Schultze. Vol. 2: Placodermi. Stuttgart: Gustav Fischer
Verlag. 128 pp.
Diaz-Martinez E, Suarez-Riglos M, Leli`
evre H, Janvier P. 1996. Premi`ere d´
ecouverte d’un Arthrodire
(Placodermi, Vertebrata) dans le D´
evonien d’Am´
erique du Sud. C. R. Acad. Sci. Paris 323:349–56
Downs JP, Donoghue PCJ. 2009. Skeletal histology of Bothriolepis canadensis (Placodermi, Antiarchi) and
evolution of the skeleton at the origin of jawed vertebrates. J. Morphol. 270:1364–80
Dupret V. 2008. A new wuttagoonaspid (Placodermi, Arthrodira) from the Lower Devonian of Yunnan (South
China): origin, dispersal, and paleobiogeographic significance. J. Vertebr. Paleontol. 28:12–20
Dupret V, Zhu M. 2008. The earliest phyllolepid (Placodermi, Arthrodira) from the Late Lochkovian (Early
Devonian) of Yunnan (South China). Geol. Mag. 145:257–78
Forey P, Janvier P. 1994. Evolution of the early vertebrates. Am. Sci. 82:554–65
Friedman M. 2007. Styloichthys as the oldest coelacanth: implications for early osteichthyan interrelationships.
J. Syst. Paleontol. 5:289–343
Goodrich ES. 1930. Studies on the Structure and Development of Vertebrates. London: Macmillan
Goujet D. 1973. Sigaspis, un nouvel arthrodire du D´
evonien inferieur du Spitsberg. Palaeontographica Abt. A
Goujet D. 1984a. Les poissons placodermes du Spitsberg. Arthrodires Dolichothoraci de la Formation de Wood Bay
(D´evonien Inf´erieur). Cahiers de Pal´
eontologie, Section Vertebr´
es. Paris: Editions du Centre National de
la Recherche Scientifique. 284 pp.
Goujet D. 1984b. Placoderm interrelationships: a new interpretation, with a short review of placoderm clas-
sifications. Proc. Linn. Soc. N. S. W. 107:211–43
Goujet D. 2001. Placoderms and basal gnathostome apomorphies. In Major Events in Early Vertebrate Evolution:
Palaeontology, Phylogeny, Genetics and Development, ed. PE Ahlberg. Syst. Assoc. Spec. Vol. 61:209–22
Goujet D, Janvier P, Suarez-Riglos M. 1984. Devonian vertebrates from South America. Nature 312:311
Goujet D, Young GC. 1995. Interrelationships of placoderms revisited. Geobios 28(Suppl. 2):89–95
Goujet D, Young GC. 2004. Placoderm anatomy and phylogeny: new insights. In Recent Advances in the Origin
and Early Radiation of Vertebrates, ed. G Arratia, MVH Wilson, R Cloutier, pp. 109–26. Munich: Verlag
Dr. Friedrich Pfeil
Gross W. 1932. Die Arthrodira Wildungens. Geol. Pal¨aontol. Abh. 19:5–61
546 Young
Annu. Rev. Earth Planet. Sci. 2010.38:523-550. Downloaded from
by Australian National University on 05/30/10. For personal use only.
EA38CH20-Young ARI 29 March 2010 14:51
Gross W. 1961. Lunaspis broilii und Lunaspis heroldi aus dem Hunsr ¨
uckschiefer (Unterdevon, Rheinland).
Notizbl. Hess. Landes. Bodenforsch. Wiesbaden 89:17–43
Gross W. 1962. Neuuntersuchung der Stensioellida (Arthrodira, Unterdevon). Notizbl. Hess. Landes. Boden-
forsch. Wiesbaden 90:48–86
Gross W. 1963. Gemuendina stuertzi Traquair. Neuuntersuchung. Notizbl. Hess. Landes. Bodenforsch. Wiesbaden
om BM, Janke A. 2009. Gnathostome phylogenomics utilizing lungfish EST sequences. Mol. Biol. Evol.
Heintz A. 1932. The structure of Dinichthys: a contribution to our knowledge of the Arthrodira. In The Bashford
Dean Memorial Volume: Archaic Fishes, ed. EW Gudger, pp. 115–224. New York: Am. Mus. Nat. Hist.
Huxley TH. 1861. Preliminary essay upon the systematic arrangement of the fishes of the Devonian epoch.
Mem. Geol. Surv. U. K. 10:1–40
Jaekel O. 1907. Ueber Pholidosteus nov. gen., und die Mundbildung und die K ¨
orperform der Placodermen.
Sber. Ges. Naturf. Freunde Berlin 1907:170–86
Janvier P. 1996. Early Vertebrates. Oxford: Clarendon. 393 pp.
Janvier P. 2007. Living primitive fishes and fishes from deep time. Fish Physiol. 26:1–51
Janvier P, Cl´
ement G. 2005. A new groenlandaspidid arthrodire (Vertebrata: Placodermi) from the Famennian
of Belgium. Geol. Belgica 8:51–67
Janvier P, Pan J. 1982. Hyrcanaspis bliecki n.g. n.sp., a new primitive euantiarch (Antiarcha, Placodermi) from
the Eifelian of northeastern Iran, with a discussion on antiarch phylogeny. N. Jb. Geol. Pal¨aontol. Abh.
Johanson Z. 2002. Vascularization of the osteostracan and antiarch (Placodermi) pectoral fin: similarities, and
implications for placoderm relationships. Lethaia 35:169–86
Johanson Z, Smith MM. 2005. Origin and evolution of gnathostome dentitions: a question of teeth and
pharyngeal denticles in placoderms. Biol. Rev. 80:303–45
Kiaer J. 1915. Upper Devonian fish remains from Ellesmere Land, with remarks on Drepanaspis.Rep. Second
Nor. Arct. Exped. “Fram” 1898–1902, No. 33
evre H, Al-Muallen MS, Goujet D, Halewani M, Janjou D, et al. 1999. Early Devonian vertebrates from
the Jauf Formation, Saudi Arabia. Saudi Arab. Deputy Ministry Miner. Res. Tech. Rep. BRGM-TR-13–5.
26 pp.
evre H, Janvier P, Blieck A. 1993. Silurian-Devonian vertebrate biostratigraphy of western Gondwana
and related terranes (South America, Armorica-Bohemia, Middle East). See Long 1993, pp. 139–73
Liu TS, P’an K. 1958. Devonian fishes from Wutung Series near Nanking, China. Palaeontol. Sinica 141(Ser.
C15):1–41 (In Chinese and English)
Liu YH. 1963. On the Antiarchi from Chutsing, Yunnan. Vertebr. PalAsiat. 7:39–46 (In Chinese with English
Liu YH. 1973. On the new forms of Polybranchiaspiformes and Petalichthyida from Devonian of southwest
China. Vertebr. PalAsiat. 11:132–43 (In Chinese with English summary)
Long JA. 1984. New phyllolepids from the Victoria, and the relationships of the group. Proc. Linn. Soc. N. S.
W. 107:263–308
Long JA, ed. 1993. Palaeozoic Vertebrate Biostratigraphy and Biogeography. London: Belhaven. 369 pp.
Long JA. 1997. Ptyctodontid fishes (Vertebrata, Placodermi) from the Late Devonian Gogo Formation,
Western Australia, with a revision of the European genus Ctenurella Ørvig, 1960. Geodiversitas 19:515–55
Long JA, Trinajstic K. 2010. The Late Devonian Gogo Formation L¨
agerstatte of Western Australia: excep-
tional early vertebrate preservation and diversity. Annu. Rev. Earth Planet. Sci. 38:255–79
Long JA, Trinajstic KM, Johanson Z. 2009. Devonian arthrodire embryos and the origin of internal fertilization
in vertebrates. Nature 457:1124–27
Long JA, Trinajstic KM, Young GC, Senden T. 2008. Live birth in the Devonian period. Nature 453:650–52
Long JA, Young GC. 1988. Acanthothoracid remains from the Early Devonian of New South Wales, including
a complete sclerotic capsule and pelvic girdle. Assoc. Australas. Palaeontol. Mem. 7:65–80
cs E. 1991. New Remigolepis (Pisces, Antiarchi) from the Famennian deposits of the Central Devonian
Field (Russia, Tula region). Daba Muz. 3:51–56 (In Russian with English abstract) Placoderms 547
Annu. Rev. Earth Planet. Sci. 2010.38:523-550. Downloaded from
by Australian National University on 05/30/10. For personal use only.
EA38CH20-Young ARI 29 March 2010 14:51
Mark-Kurik E. 2000. The Middle Devonian fishes of the Baltic States (Estonia, Latvia) and Belarus. Cour.
Forsch. Inst. Senckenberg 223:309–24
Miles RS. 1967. Observations on the ptyctodont fish, Rhamphodopsis Watson. Zool. J. Linn. Soc. 47:99–120
Miles RS. 1969. Features of placoderm diversification and the evolution of the arthrodire feeding mechanism.
Trans. R. Soc. Edinb. 68:123–70
Miles RS. 1971. The Holonematidae (placoderm fishes): a review based on new specimens of Holonema from
the Upper Devonian of Western Australia. Phil.Trans.R.Soc.B.263:101–234
Miles RS, Dennis K. 1979. A primitive eubrachythoracid arthrodire from Gogo, Western Australia. Zool. J.
Linn. Soc. 66:31–62
Miles RS, Westoll TS. 1968. The placoderm fish Coccosteus cuspidatus Miller ex Agassiz from the Middle Old
Red Sandstone of Scotland. Part 1: Descriptive morphology. Trans. R. Soc. Edinb. 67:373–476
Miles RS, Young GC. 1977. Placoderm interrelationships reconsidered in the light of new ptyctodontids from
Gogo Western Australia. Linn. Soc. Symp. Ser. 4:123–98
Miller H. 1841. The Old Red Sandstone. Edinburgh: Adam and Charles Black. 275 pp.
Newman MJ, Trewin NG. 2008. Discovery of the arthrodire genus Actinolepis (class Placodermi) in the Middle
Devonian of Scotland. Scott. J. Geol. 44:83–88
Obruchev DV. 1964. Osnovy Paleontologii. Vol. 11: Beschelyostni i Ribi. Moscow: Nauka. Engl. transl. 1967.
Fundamentals of Paleontology, Agnatha & Pisces. Isr. Program Sci. Transl.
Pan J, Huo F, Cao J, Gu Q, Liu S, et al. 1987. Continental Devonian System of Ningxia and its Biotas. Beijing:
Geological Publishing House. 237 pp. (In Chinese with English summary)
Patterson C. 1977. Cartilage bones, dermal bones and membrane bones, or the exoskeleton versus the en-
doskeleton. Linn. Soc. Symp. Ser. 4:77–121
Reif W-E. 1982. Evolution of dermal skeleton and dentition in vertebrates: the odontode regulation theory.
Evol. Biol. 15:287–368
Reif W-E. 2002. Evolution of the dermal skeleton of vertebrates: concepts and methods. Neues Jahrb. Geol.
Pal¨aontol. Abh. 223:53–78
Ritchie A. 1973. Wuttagoonaspis gen. nov., an unusual arthrodire from the Devonian of Western New South
Wales, Australia. Palaeontogr. Abt. A 143:58–72
Ritchie A. 1984. A new placoderm, Placolepis gen. nov. (Phyllolepidae) from the Late Devonian of New South
Wales, Australia. Proc. Linn. Soc. N. S. W. 107:321–53
Ritchie A. 2005. Cowralepis, a new genus of phyllolepid fish (Pisces, Placodermi) from the Late Middle Devonian
of New South Wales, Australia. Proc. Linn. Soc. N. S. W. 126:215–59
Schultze H-P, Cloutier R, eds. 1996. Devonian Fishes and Plants of Miguasha, Quebec, Canada. Munich: Verlag
Dr. Friedrich Pfeil
Schultze H-P, Cumbaa SL. 2001. Dialipina and the characters of basal actinopterygians. In Major Events in
Early Vertebrate Evolution: Palaeontology, Phylogeny, Genetics and Development, ed. PE Ahlberg. Syst. Assoc.
Spec. Vol. 61:315–32
Sedgewick A, Murchison RI. 1828. On the structure and relations of the deposits contained between the
Primary Rocks and the Oolitic Series in the north of Scotland. Trans. Geol. Soc. London 3:125–60
Sedgewick A, Murchison RI. 1839. On the classification of the older rocks of Devonshire and Cornwall.
Proc. Geol. Soc. London 3:121–23
Sepkoski JJ. 2002. A compendium of fossil marine animal genera. Bull. Am. Paleontol. 363:1–560
o EA. 1925. On the head of the macropetalichthyids, with certain remarks on the head of the other
arthrodires. Publ. Field Mus. Nat. Hist. (Geol.) 4:87–197
o EA. 1931. Upper Devonian vertebrates from East Greenland, collected by the Danish Greenland
expedition in 1929 and 1930. Medd. Grønl. 86:1–212
o EA. 1934. On the Placodermi of the Upper Devonian of East Greenland. 1: Phyllolepida and
Arthrodira. Medd. Grønl. 97:1–58
o EA. 1948. On the Placodermi of the Upper Devonian of East Greenland. 2: Antiarchi: subfamily
Bothriolepinae. With an attempt at a revision of the previously described species of that subfamily. Medd.
om Grønland 139(Palaeozool. Groenl. 2):1–622
o EA. 1959. On the pectoral fin and shoulder girdle of the arthrodires. Kungl. Svensk.VetenskAkad. Handl.
548 Young
Annu. Rev. Earth Planet. Sci. 2010.38:523-550. Downloaded from
by Australian National University on 05/30/10. For personal use only.
EA38CH20-Young ARI 29 March 2010 14:51
o EA. 1963. Anatomical studies of the arthrodiran head. Part 1. Kungl. Svensk.VetenskAkad. Handl.
o EA. 1969. Elasmobranchiomorphi Placodermata Arthrodires. In Trait´e de Pal´eontologie, ed. J Piveteau,
4(2):71–692. Paris: Masson
Trinajstic K, Dennis-Bryan K. 2009. Phenotypic plasticity, polymorphism and phylogeny within placoderms.
Acta Zool. 90:83–102
Wang HC. 1943. The stratigraphic position of the Devonian fish-bearing series (Bothriolepis bed) of eastern
Yunnan. Sci. Rec. 1:562–64
Watson DMS. 1938. On Rhamphodopsis, a ptyctodont from the Middle Old Red Sandstone of Scotland.
Trans. R. Soc. Edinb. 59:397–410
Westoll TS. 1958. The origin of continental vertebrate faunas. Trans. Geol. Soc. Glasg. 23;1–27
Westoll TS. 1979. Devonian fish biostratigraphy. Spec. Pap. Palaeontol. 23:341–53
White EI. 1952. Australian arthrodires. Bull. Br. Mus. (Nat. Hist.) (Geol.) 1:249–304
White EI. 1969. The deepest vertebrate fossil and other arctolepid fishes. Biol. J. Linn. Soc. London 1:293–310
White EI. 1978. The larger arthrodiran fishes from the area of the Burrinjuck Dam, N. S. W. Trans. Zool. Soc.
London 34:149–262
Whitfield JB, Lockhart PJ. 2007. Deciphering rapid radiations. Trends Ecol. Evol. 22:258–65
Woodward AS. 1891. Catalogue of the Fossil Fishes in the British Museum (Natural History). Part II. London:
Brit. Mus. (Nat. Hist.)
Woodward AS. 1915. Preliminary report on the fishes from Dura Den. Rep. Br. Assoc. Adv. Sci. 84(1914):122–23
Woodward AS. 1916. On Devonian fish remains from Australia and the Antarctic regions. Proc. Geol. Soc.
London 72:79–80
Young GC. 1979. New information on the structure and relationships of Buchanosteus (Placodermi, Euar-
throdira) from the Early Devonian of New South Wales. Zool. J. Linn. Soc. 66:309–52
Young GC. 1980. A new Early Devonian placoderm from New South Wales, Australia, with a discussion of
placoderm phylogeny. Palaeontogr. Abt. A 167:10–76
Young GC. 1981. New Early Devonian brachythoracids (placoderm fishes) from the Taemas–Wee Jasper
region of New South Wales. Alcheringa 5:245–71
Young GC. 1986. The relationships of placoderm fishes. Zool. J. Linn. Soc. 88:1–57
Young GC. 1988. Antiarchs (placoderm fishes) from the Devonian Aztec Siltstone, southern Victoria Land,
Antarctica. Palaeontogr. Abt. A 202:1–125
Young GC. 1991. Fossil fishes from Antarctica. In The Geology of Antarctica, ed. RJ Tingey, pp. 538–67. Oxford:
Oxford Univ. Press
Young GC. 1999. Preliminary report on the biostratigraphy of new placoderm discoveries in the Hervey
Group (Upper Devonian) of central New South Wales. Rec. West. Aust. Mus. Suppl. 57:139–50
Young GC. 2003a. Did placoderm fish have teeth? J. Vertebr. Paleontol. 23:987–90
Young GC. 2003b. North Gondwanan mid-Palaeozoic connections with Euramerica and Asia: Devonian
vertebrate evidence. Cour. Forsch. Inst. Senckenberg 242:169–85
Young GC. 2004. Large brachythoracid arthrodires (placoderm fishes) from the Early Devonian of Wee
Jasper, New South Wales, Australia. J. Vertebr. Paleontol. 24:1–17
Young GC. 2005a. An articulated phyllolepid fish (Placodermi) from the Devonian of central Australia: im-
plications for non-marine connections with the Old Red Sandstone continent. Geol. Mag. 142:173–86
Young GC. 2005b. New phyllolepids (placoderm fishes) from the Middle-Late Devonian of southeastern
Australia. J. Vertebr. Paleontol. 25:261–73
Young GC. 2005c. A new phyllolepid placoderm occurrence (Devonian fish) from the Dulcie Sandstone,
Georgina Basin, central Australia. Proc. Linn. Soc. N. S. W. 126:203–13
Young GC. 2006. Biostratigraphic and biogeographic context for tetrapod origins during the Devonian:
Australian evidence. Alcheringa 30(Suppl. 1):409–28
Young GC. 2007. Devonian formations, vertebrate faunas, and age control on the far south coast of New
South Wales and adjacent Victoria. Aust. J. Earth Sci. 54:991–1008
Young GC. 2008a. Number and arrangement of extraocular muscles in primitive gnathostomes: evidence
from extinct placoderm fishes. Biol. Lett. 4:110–14 Placoderms 549
Annu. Rev. Earth Planet. Sci. 2010.38:523-550. Downloaded from
by Australian National University on 05/30/10. For personal use only.
EA38CH20-Young ARI 29 March 2010 14:51
Young GC. 2008b. Early evolution of the vertebrate eye—fossil evidence. Evol. Educ. Outreach 1(4):427–38
Young GC. 2008c. The relationships of antiarchs (Devonian placoderm fishes)—evidence supporting placo-
derm monophyly. J. Vertebr. Paleontol. 28:626–36
Young GC. 2009. New arthrodires (Family Williamsaspididae) from Wee Jasper, New South Wales (Early
Devonian), with comments on placoderm morphology and palaeoecology. Acta Zool. 90(Suppl. 1):69–82
Young GC, Burrow CJ, Long JA, Turner S, Choo B. 2010. Devonian macrovertebrate assem-
blages and biogeography of East Gondwana (Australasia, Antarctica). Palaeoworld. In press,
Young GC, Goujet D. 2003. Devonian fish remains from the Dulcie Sandstone and Cravens Peak Beds,
Georgina Basin, central Australia. Rec. West. Aust. Mus. Suppl. 65:1–85
Young GC, Leli`
evre H, Goujet D. 2001. Primitive jaw structure in an articulated brachythoracid arthrodire
(placoderm fish; Early Devonian) from southeastern Australia. J. Vertebr. Paleontol. 18:261–74
Young GC, Long JA. 2005. Phyllolepid placoderm fish remains from the Devonian Aztec Siltstone, southern
Victoria Land, Antarctica. Antarct. Sci. 17:387–408
Young GC, Zhang GR. 1992. Structure and function of the pectoral joint and operculum in antiarchs, Devonian
placoderm fishes. Palaeontology 35:443–64
Young GC, Zhang GR. 1996. New information on the morphology of yunnanolepid antiarchs (placoderm
fishes) from the Early Devonian of South China. J. Vertebr. Paleontol. 16:623–41
Zalc B, Goujet D, Colman D. 2007. The origin of the myelination program in vertebrates. Curr. Biol. 18:R511–
Zhang GR. 1978. The antiarchs from the Early Devonian of Yunnan. Vertebr. PalAsiat. 16:147–86 (In Chinese
with English summary)
Zhang GR, Wang JQ, Wang ST. 2001. The structure of pectoral fin and tail of Yunnanolepidoidei, with a
discussion of the pectoral fin of chuchinolepids. Vertebr. PalAsiat. 39:1–13 (In Chinese and English)
Zhao WJ, Zhu M. 2007. Diversification and faunal shift of Siluro-Devonian vertebrates of China. Geol. J.
Zhu M. 1991. New information on Diandongpetalichthys (Placodermi: Petalichthyida). In Early Vertebrates and
Related Problems of Evolutionary Biology, ed. Chang MM, Liu YH, Zhang GR, pp. 179–94. Beijing: Science
Zhu M. 1996. The phylogeny of the Antiarcha (Placodermi, Pisces), with a description of Early Devonian
antiarchs from Qujing, Yunnan, China. Bull. Mus. Natl. Hist. Nat. 18:233–347
Zhu M. 2000. Catalogue of Devonian vertebrates in China, with notes on bio-events. Cour. Forsch. Inst.
Senckenberg 223:373–90
Zhu M, Ahlberg PE, Zhao W, Jia L. 2002. First Devonian tetrapod from Asia. Nature 420:760–61
Zhu M, Janvier P. 1996. A small antiarch, Minicrania lirouyii gen. et sp. nov., from the Early Devonian of
Qujing, Yunnan (China), with remarks on antiarch phylogeny. J. Vertebr. Paleontol. 16:1–15
Zhu M, Schultze H-P. 2001. Interrelationships of basal osteichthyans. In Major Events in Early Vertebrate
Evolution: Palaeontology, Phylogeny, Genetics and Development, ed. PE Ahlberg. Syst. Assoc. Spec. Vol. 61:289–
Zhu M, Wang J-Q. 2000. Silurian vertebrate assemblages of China. Cour. Forsch. Inst. Senckenberg 223:161–68
Zhu M, Wang N-Z, Wang J-Q. 2000. Devonian macro- and microvertebrate assemblages of China. Cour.
Forsch. Inst. Senckenberg 223:361–72
Zhu M, Yu XB. 2004. Lower jaw character transitions among major sarcopterygian groups—a survey based
on new materials from Yunnan, China. In Recent Advances in the Origin and Early Radiation of Vertebrates,
ed. G Arratia, MVH Wilson, R Cloutier, pp. 271–86. Munich: Verlag Dr. Friedrich Pfeil
Zhu M, Yu XB, Janvier P. 1999. A primitive fish sheds light on the origin of bony fishes. Nature 397:607–10
Zhu M, Yu XB, Wang W, Zhao WJ, Jia LT. 2006. A primitive fish provides key characters bearing on deep
osteichthyan phylogeny. Nature 441:77–80
Zhu M, Zhao WJ, Jia LT, Lu J, Qiao T, Qu QM. 2009. The oldest articulated osteichthyan reveals mosaic
gnathostome characters. Nature 458:469–74
550 Young
Annu. Rev. Earth Planet. Sci. 2010.38:523-550. Downloaded from
by Australian National University on 05/30/10. For personal use only.
AR409-FM ARI 29 March 2010 12:13
Annual Review
of Earth and
Planetary Sciences
Volume 38, 2010 Contents
Ikuo Kushiro ppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppxiv
Toward the Development of “Magmatology”
Ikuo Kushiro pppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppp1
Nature and Climate Effects of Individual Tropospheric Aerosol
Mih´aly P´osfai and Peter R. Buseck ppppppppppppppppppppppppppppppppppppppppppppppppppppppppppp17
The Hellenic Subduction System: High-Pressure Metamorphism,
Exhumation, Normal Faulting, and Large-Scale Extension
Uwe Ring, Johannes Glodny, Thomas Will, and Stuart Thomson ppppppppppppppppppppppppp45
Orographic Controls on Climate and Paleoclimate of Asia: Thermal
and Mechanical Roles for the Tibetan Plateau
Peter Molnar, William R. Boos, and David S. Battisti pppppppppppppppppppppppppppppppppppppp77
Lessons Learned from the 2004 Sumatra-Andaman
Megathrust Rupture
Peter Shearer and Roland B¨urgmann pppppppppppppppppppppppppppppppppppppppppppppppppppppp103
Oceanic Island Basalts and Mantle Plumes: The Geochemical
William M. White pppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppp133
Isoscapes: Spatial Pattern in Isotopic Biogeochemistry
Gabriel J. Bowen pppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppp161
The Origin(s) of Whales
Mark D. Uhen pppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppp189
Frictional Melting Processes in Planetary Materials:
From Hypervelocity Impact to Earthquakes
John G. Spray ppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppp221
The Late Devonian Gogo Formation L¨
agerstatte of Western Australia:
Exceptional Early Vertebrate Preservation and Diversity
John A. Long and Kate Trinajstic pppppppppppppppppppppppppppppppppppppppppppppppppppppppppp255
Annu. Rev. Earth Planet. Sci. 2010.38:523-550. Downloaded from
by Australian National University on 05/30/10. For personal use only.
AR409-FM ARI 29 March 2010 12:13
Booming Sand Dunes
Melany L. Hunt and Nathalie M. Vriend ppppppppppppppppppppppppppppppppppppppppppppppppp281
The Formation of Martian River Valleys by Impacts
Owen B. Toon, Teresa Segura, and Kevin Zahnle ppppppppppppppppppppppppppppppppppppppppp303
The Miocene-to-Present Kinematic Evolution of the Eastern
Mediterranean and Middle East and Its Implications for Dynamics
Xavier Le Pichon and Corn´e Kreemer pppppppppppppppppppppppppppppppppppppppppppppppppppppp323
Oblique, High-Angle, Listric-Reverse Faulting and Associated
Development of Strain: The Wenchuan Earthquake of May 12,
2008, Sichuan, China
Pei-Zhen Zhang, Xue-ze Wen, Zheng-Kang Shen, and Jiu-hui Chen pppppppppppppppppp353
Composition, Structure, Dynamics, and Evolution of Saturn’s Rings
Larry W. Esposito pppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppp383
Late Neogene Erosion of the Alps: A Climate Driver?
Sean D. Willett pppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppp411
Length and Timescales of Rift Faulting and Magma Intrusion:
The Afar Rifting Cycle from 2005 to Present
Cynthia Ebinger, Atalay Ayele, Derek Keir, Julie Rowland, Gezahegn Yirgu,
Tim Wright, Manahloh Belachew, and Ian Hamling ppppppppppppppppppppppppppppppppppp439
Glacial Earthquakes in Greenland and Antarctica
Meredith Nettles and G¨oran Ekstr¨om pppppppppppppppppppppppppppppppppppppppppppppppppppppp467
Forming Planetesimals in Solar and Extrasolar Nebulae
E. Chiang and A.N. Youdin ppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppp493
Placoderms (Armored Fish): Dominant Vertebrates
of the Devonian Period
Gavin C. Young pppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppp523
The Lithosphere-Asthenosphere Boundary
Karen M. Fischer, Heather A. Ford, David L. Abt, and Catherine A. Rychert pppppppppp551
Cumulative Index of Contributing Authors, Volumes 28–38 ppppppppppppppppppppppppppp577
Cumulative Index of Chapter Titles, Volumes 28–38 pppppppppppppppppppppppppppppppppppp581
An online log of corrections to Annual Review of Earth and Planetary Sciences articles
may be found at
Contents ix
Annu. Rev. Earth Planet. Sci. 2010.38:523-550. Downloaded from
by Australian National University on 05/30/10. For personal use only.
... Les placodermes étaient le groupe de vertébrés le plus diversifié durant le Dévonien, il y a 419 à 359 millions d'années (Denison, 1978;Janvier, 1996;Young, 2010;Long, 2011). ...
... Des plaques d'os dermiques forment l'armure autour de la tête et du thorax des espèces de ce groupe complètement éteint depuis plus de 350 millions d'années (Young, 2010). Ces poissons avaient une distribution cosmopolite durant le Dévonien et étaient présents dans les environnements d'eaux douces et marins de manière analogue aux poissons téléostéens actuels (Young, 2010). ...
... Des plaques d'os dermiques forment l'armure autour de la tête et du thorax des espèces de ce groupe complètement éteint depuis plus de 350 millions d'années (Young, 2010). Ces poissons avaient une distribution cosmopolite durant le Dévonien et étaient présents dans les environnements d'eaux douces et marins de manière analogue aux poissons téléostéens actuels (Young, 2010). Les placodermes incluaient le plus grand vertébré de l'époque [Titanichthys avec une tête d'environ 1 m de long (Boyle et al., 2017)] ainsi que le plus petit [Minicrania avec une tête de moins de 2,2 mm (Zhu et al., 1996;Young, 2010)]. ...
Full-text available
Groenlandaspis représente un genre d’arthrodire largement répandu sur la planète au Dévonien. Ils sont retrouvés au Gondwana à partir du Praguien et n’étaient retrouvés en Euramérique qu’à partir du Famennien. Une nouvelle espèce de Groenlandaspis est décrite provenant du Givétien de Cairo, New York, États-Unis. Cette espèce est caractérisée par une plaque latérale postérieure allongée en hauteur et amincie sur la partie supérieure; une plaque antérieure ventrolatérale possédant une large zone de chevauchement pour la plaque interlaterale, une marge latérale droite (zone de chevauchement pour la plaque spinale) et une absence d’extension latérale ainsi qu’une plaque antérieure latérale avec une marge dorsale relativement droite avec une entaille pectorale faiblement définie. La nouvelle espèce est l’occurrence la plus ancienne en Amérique du Nord ainsi qu’en Euramérique. Cette occurrence supporte le point de vue que ce genre était beaucoup plus répandu environ 10 millions d’années plus tôt que l’on ne le croyait précédemment. Cette évidence pointe vers un cosmopolitisme de la famille Groenlandispididae et du genre Groenlandaspis tôt durant le milieu du Dévonien tel qu’il a déjà été observé chez d’autres espèces de chondrichtyens par exemple. La première phylogénie de la famille est également présentée. L’arbre de consensus strict suggère la paraphylie des Groenlandaspididae et la polyphylie de Groenlandaspis.
... The dermatoskeleton of Xiushanosteus displays an intriguing mélange of characters from various placoderm subgroups. The skull roof profile with parallel lateral margins and deeply embayed posterior margin resembles that of acanthothoracids such as Romundina [14][15][16][17] (Extended Data Fig. 7a,b). However, the pattern of the skull roof plates generally resembles that of actinolepidoid arthrodires, possessing only one pair of postorbitals and paranuchals. ...
... However, the pattern of the skull roof plates generally resembles that of actinolepidoid arthrodires, possessing only one pair of postorbitals and paranuchals. The main lateral lines of the skull roof resemble those of maxillate placoderms 10,12,20,21 and acanthothoracids [14][15][16][17] in having gently curved and sub-parallel trajectories, rather than trajectories with a 'dog-leg' deflection as seen in, for example, petalichthyids 22 and some nested arthrodires 23 . The nuchal does not extend anteriorly to separate the central plates, in contrast to the condition found in acanthothoracids, petalichthyids, and antarctaspidid and wuttagoonaspid arthrodires 16 . ...
... The main lateral lines of the skull roof resemble those of maxillate placoderms 10,12,20,21 and acanthothoracids [14][15][16][17] in having gently curved and sub-parallel trajectories, rather than trajectories with a 'dog-leg' deflection as seen in, for example, petalichthyids 22 and some nested arthrodires 23 . The nuchal does not extend anteriorly to separate the central plates, in contrast to the condition found in acanthothoracids, petalichthyids, and antarctaspidid and wuttagoonaspid arthrodires 16 . The trunk shield of Xiushanosteus is short, resembling the acanthothracid Romundina 15 and the petalichthyid Lunaspis 24 . ...
Full-text available
Molecular studies suggest that the origin of jawed vertebrates was no later than the Late Ordovician period (around 450 million years ago (Ma))1,2. Together with disarticulated micro-remains of putative chondrichthyans from the Ordovician and early Silurian period3–8, these analyses suggest an evolutionary proliferation of jawed vertebrates before, and immediately after, the end-Ordovician mass extinction. However, until now, the earliest complete fossils of jawed fishes for which a detailed reconstruction of their morphology was possible came from late Silurian assemblages (about 425 Ma)9–13. The dearth of articulated, whole-body fossils from before the late Silurian has long rendered the earliest history of jawed vertebrates obscure. Here we report a newly discovered Konservat-Lagerstätte, which is marked by the presence of diverse, well-preserved jawed fishes with complete bodies, from the early Silurian (Telychian age, around 436 Ma) of Chongqing, South China. The dominant species, a ‘placoderm’ or jawed stem gnathostome, which we name Xiushanosteus mirabilis gen. et sp. nov., combines characters from major placoderm subgroups14–17 and foreshadows the transformation of the skull roof pattern from the placoderm to the osteichthyan condition¹⁰. The chondrichthyan Shenacanthus vermiformis gen. et sp. nov. exhibits extensive thoracic armour plates that were previously unknown in this lineage, and include a large median dorsal plate as in placoderms14–16, combined with a conventional chondrichthyan bauplan18,19. Together, these species reveal a previously unseen diversification of jawed vertebrates in the early Silurian, and provide detailed insights into the whole-body morphology of the jawed vertebrates of this period.
... Placodermi is an extinct group of jawed vertebrates that first occurred in the Silurian, then dominated the Devonian and constituted a prevalent biotic component of the marine vertebrate ecosystem from 425.0 to 358.9 million years ago (Carr, 1995;Denison, 1978;Janvier, 1996;Young, 2010;Zhu, 1996). Recent prevailing phylogenetic hypotheses placed Placodermi as jawed stem-Gnathostomata that is sister to crown-Gnathostomata or modern jawed vertebrates (Brazeau, 2009;Davis et al., 2012;Dupret et al., 2014;Giles et al., 2015;King, 2021;Long et al., 2015;Qiao et al., 2016;Trinajstic et al., 2015;Zhu et al., 2013Zhu et al., , 2016. ...
... Silhouettes indicate groups of Antiarcha. 2011; Young, 2010), Antiarcha has contributed significantly to the Devonian stratigraphic correlation. For instance, the biozonation of the East Baltic and southern East Antarctica Devonian succession is partly based upon the antiarchs Bothriolepis, Asterolepis, and Pambulaspis (Young, 1974(Young, , 1988. ...
Full-text available
Antiarcha data are essential to quantitative studies of basal jawed vertebrates. The absence of struc-tured data on key groups of early vertebrates, such as Antiarcha, has lagged in understanding their diversity and distribution patterns. Previous works of early vertebrates usually focused on anatomy and phylogeny, given their significant impacts on the evolution of key characters, but lacked comprehensive structured data. Here, we contribute an unprecedented open-access Antiarcha dataset covering 60 genera of 6025 specimens from the Ludfor-dian to the Famennian globally. We have organized an expert team to collect and curate 142 publications spanning from 1939 to 2021. Additionally, we have two-stage quality controls in the process: domain experts examined the literature and senior experts reviewed the results. In this paper, we give details of the data storage structure and visualize these antiarch fossil sites on the paleogeographic map. The novel Antiarcha dataset has tremendous research potential, including testing previous qualitative hypotheses in biodiversity changes, spatiotemporal distribution , evolution, and community composition. It is now an essential part of the DeepBone database and will be updated with the latest publication, also available on (Pan and Zhu, 2021).
... A framework for testing competition in the fossil record should therefore include data independent of stratigraphy and paleogeography. Agnathan classes considered in this study (Anaspida, Heterostraci, Osteostraci, and Thelodonti) overlap stratigraphically (Dineley and Loeffler 1993;Janvier and Blieck 1993;Zhu et al. 2009;Lu et al. 2016;Supplementary Fig. S1) and geographically (specifically Euramerica; Dineley and Loeffler 1993;Young 1993;Newman and Trewin 2001) with each of the gnathostome classes (Acanthodii, Actinopterygii, Chondrichthyes, Placodermi, and Sarcopterygii), but not necessarily at finer taxonomic levels or at individual sites; therefore, we limit our comparisons to among classes, with the exception of the highly disparate orders of Placodermi (Young 2010) and the Furcaudiformes, a morphologically distinct order within Thelodonti (Wilson and Caldwell 1998). When potentially competing taxa are identified based on our functional-morphological criteria, we can use what stratigraphic or geographic data we have to further support or reject these hypotheses. ...
... Anaspida, Osteostraci, and Thelodonti may have competed with gnathostomes, as all three agnathan groups overlap stratigraphically and geographically with gnathostome taxa (Dineley and Loeffler 1993;Janvier and Blieck 1993;Newman and Trewin 2001;Young 2010) and are in the same morphospace on axes 1 and 2 (Fig. 2). However, body forms across the region are highly variable, and no specific gnathostome group is more similar to agnathans than any other group (Fig. 3), and even similar taxa may be sufficiently different to avoid competition with one another. ...
Full-text available
The rise of jawed vertebrates (gnathostomes) and extinction of nearly all jawless vertebrates (agnathans) is one of the most important transitions in vertebrate evolution, but the causes are poorly understood. Competition between agnathans and gnathostomes during the Devonian period is the most commonly hypothesized cause; however, no formal attempts to test this hypothesis have been made. Generally, competition between species increases as morphological similarity increases; therefore, this study uses the largest to date morphometric comparison of Silurian and Devonian agnathan and gnathostome groups to determine which groups were most and least likely to have competed. Five agnathan groups (Anaspida, Heterostraci, Osteostraci, Thelodonti, and Furcacaudiformes) were compared with five gnathostome groups (Acanthodii, Actinopterygii, Chondrichthyes, Placodermi, and Sarcopterygii) including taxa from most major orders. Morphological dissimilarity was measured by Gower's dissimilarity coefficient, and the differences between agnathan and gnathostome body forms across early vertebrate morphospace were compared using principal coordinate analysis. Our results indicate competition between some agnathans and gnathostomes is plausible, but not all agnathan groups were similar to gnathostomes. Furcacaudiformes (fork-tailed thelodonts) are distinct from other early vertebrate groups and the least likely to have competed with other groups.
... Antiarch placoderm fishes are probably the most abundant middle Paleozoic vertebrates, known from almost all paleozoogeographic provinces, from the Silurian and Devonian periods (Lebedev and Zakharenko, 2010;Young, 2010). Many Devonian localities have yielded completely preserved skeletons. ...
... During the Emsian, the antiarchs expanded to Gondwana, and at the beginning of the Eifelian, to Euramerica (Young, 1990(Young, , 2010Young et al., 2010) (Fig. 14.1). The extinction of the cyathaspidiform and most of the pteraspidiform agnathans by the Eifelian in Euramerica coincided with the antiarch invasion to these new territories (Fig. 14.2) rich in resources. ...
Antiarch placoderm fishes were an abundant component of the Middle Paleozoic vertebrate assemblages. Despite a large number of known taxa and specimens, the morphology and function of the skeletal elements of their jaws is inadequately known. Because of this, questions regarding their feeding modes and their roles in the trophic webs remain open. We present a skeletomuscular model of the antiarch jaw apparatus with an attempt to reconstruct its potential biomechanical function. The position of the upper jaw suborbital bones within the plane of the ventral side of the fish armor is suggested to represent the natural “mouth closed” position. During mouth opening, the suborbitals rotated rostrally with simultaneous depression and inward rotation of the infragnathals. The ball-and-socket jaw articulation might ensure this combined movement. Recently described lower jaw elements of Livnolepis zadonica (Obrucheva, 1983) and Bothriolepis sp. from the Upper Devonian (lower Famennian) of Central Russia demonstrating very deep and porous blades of the oral division of the infragnathals queried the structure of these bones in other antiarchs. Observed porosity reflects intense vascularization to supply blood to a connective tissue underlying a supposed keratinous sheath, which protected and strengthened the jaws, as well as made possible scraping tough food objects, such as thallus algae, from the substrate. Having evolved during the Silurian in the Pan-Cathaysian zoogeographical province, antiarchs migrated to Gondwana during the Emsian and later to Euramerica during the Eifelian. Supposedly, antiarchs became the first macrophytophagous vertebrates occupying the trophic level of primary consumers during the late Silurian–Early Devonian. This event diversified the only previously existing predator–prey interrelationships between filter-feeding agnathans and predatory gnathostomes.
... Placoderms play a key role in the early evolution of jawed fish (Jarvik, 1980;Reif, 1982;Janvier, 1984;Young, 1986Young, , 2010Forey and Janvier, 1993;Philippe, 1996;Donoghue et al., 2000;Sansom et al., 2005;Anderson et al., 2011;Kuratani, 2012;Zhu et al., 2013Zhu et al., , 2021Friedman, 2014, 2015;Giles et al., 2015;King et al., 2017). In particular, the origins of jaw elements and teeth were intensely studied in the past decades through the Placodermi group (Smith and Johanson, 2003;Rücklin et al., 2012Rücklin et al., , 2014Zhu et al., 2013Zhu et al., , 2016Coatham et al., 2020;Vaškaninová et al., 2020;Jobbins et al., 2021). ...
Full-text available
Placoderms are an extinct group of early jawed vertebrates that play a key role in understanding the evolution of the gnathostome body plan, including the origin of novelties such as jaws, teeth, and pelvic fins. As placoderms have a poorly ossified axial skeleton, preservation of the mainly cartilaginous axial and fin elements is extremely rare, contrary to the heavily mineralized bones of the skull and thoracic armor. Therefore, the gross anatomy of the animals and body shape is only known from a few taxa, and reconstructions of the swimming function and ecology are speculative. Here, we describe articulated specimens preserving skull roofs, shoulder girdles, most fins, and body outlines of a newly derived arthrodire. Specimens of the selenosteid Amazichthys trinajsticae gen. et sp. nov. display a skull roof with reticular ornamentation and raised sensory lines like Driscollaspis, a median dorsal plate with a unique sharp posterior depression, the pelvic girdle, the proportions and shape of the pectoral, dorsal, and caudal fins as well as a laterally enlarged region resembling the lateral keel of a few modern sharks and bony fishes. Our new phylogenetic analyses support the monophyly of the selenosteid family and place the new genus in a clade with Melanosteus, Enseosteus, Walterosteus, and Draconichthys. The shape of its body and heterocercal caudal fin in combination with the pronounced "lateral keel" suggest Amazichthys trinajsticae was an active macropelagic swimmer capable of reaching high swimming speeds.
... In this regard, relevant data from placoderms would include developmental allometry and ontogenetic morphological changes of their dermal bony plates. The multiple bony plates covering the placoderm head and trunk have great potential for fossilization (Young, 2010), with varied morphologies and overlap relationships. Few allometric growth patterns have been documented in antiarch placoderms, mainly based on the genus Bothriolepis Werdelin and Long, 1986;Downs et al., 2011). ...
During fish growth, short periods of rapid changes (morphological, behavioral, physiological, ecological), or thresholds, are interspersed with longer periods of slower development, the steps. This growth pattern is known as saltatory ontogeny. Thresholds delimit main developmental stages (embryo, larva, juvenile, adult) and are periods where modifications can lead to new life-histories, and ultimately to evolutionary novelties. We sought to determine whether saltatory ontogeny could be recognized in a jawed stem-gnathostome, the Late Devonian antiarch placoderm Bothriolepis canadensis, using an extensive size series (220 specimens: 5–220 mm in armor length). The small specimens of this series reveal a previously undocumented immature feature, a preorbital depression on the premedian and lateral plates of the dermal headshield. This depression is a plesiomorphic condition reputed to be absent in highly nested antiarchs such as B. canadensis, which possess instead a preorbital recess. Our objectives were: to describe the ontogenetic morphological changes of the preorbital area in B. canadensis, including the timing of disappearance of the depression using binomial logistic regressions, and to quantify growth allometries of the premedian and lateral plates, with linear and segmented regressions. We found significant segmented allometric patterns in the premedian plate, suggesting saltatory ontogeny in B. canadensis. Moreover, segmented patterns were congruent with the ontogenetic loss of the preorbital depression. An ecomorphological hypothesis is proposed to explain these simultaneous morphological and morphometric changes in B. canadensis. A similar hypothesis is extrapolated to interpret the innovation of the preorbital recess and loss of the preorbital depression during the antiarch phylogeny.
... Moreover, Sinolepidoidei was endemic in China and Australia (Ritchie et al., 1992). ''Bothriolepidoidei" and Asterolepidoidei thrived from the Emsian and became cosmopolitan in late Devonian (Denison, 1978;Young, 2010aYoung, , 2010b. Above all, it is possible to implement the network analysis on Antiarcha. ...
Although many hypotheses have been proposed on the biogeographic evolution of early vertebrates in Silurian and Devonian, it is still difficult to reach a consensus. is a specimen-based online database hosting data of vertebrate paleontology and paleoanthropology. Here we use comprehensive records of stem gnathostomes in DeepBone to analyse the historical biogragraphy of antiarchs, a primitive group of jawed stem-Gnathostomata. We propose a network analysis-based approach to help quantify and identify pivotal biogeographic provinces. We first segmented biogeographic provinces following the conventional hypothesis and then constructed a directed-weighted temporal-biogeographic network using the Bray-Curtis dissimilarity. Finally, we proposed Evolutionary Importance (EI), a novel algorithm to identify the pivotal province, which has a substantial impact on the constituent members of the provinces in the following age quantitatively. Six biogeographic provinces were evaluated including East Gondwana, Pan-Cathaysia, Central Asia, Baltica, Siberia, and Laurentia. The EI score analysis reveals that the Pan-Cathaysia in Lochkovian is the most crucial province for antiarchs. The Siberia, Baltica, and Central Asia provinces in Givetian and Frasnian are essential to the dispersal of antiarchs, whereas the Central Asia province might act as a ‘stepping-stone’ for the dispersal of euantiarchs. Yunnanolepidoids, which are the most plesiomorphic antiarchs, were endemic in the Pan-Cathaysia, highlighting the origin of Antiarcha in this province. Sinolepidoids also originated in the Pan-Cathaysia and were endemic in the Pan-Cathaysia and East Gondwana provinces. Among euantiarchs, bothriolepidoids originated and differentiated in the Pan-Cathaysia province, while asterolepidoids originated and differentiated in the East Gondwana province.
... Each of the obtained trees was time calibrated 100 times using the R package 'paleotree' 63 by randomizing the tip age of every species within the chronostratigraphic unit, at age or subperiod rank, where their first appearance occurs. A minimum age constraint was set in the ancestral nodes of the main clades of stem-gnathostomes considering their first appearance in the fossil record (Conodonta, Furongian 64 ; Anaspida, Llandovery 20 ; Pteraspidomorphi, Darriwilian-Sandbian 65,66 ; Thelodonti, Sandbian 26 ; Osteostraci, Aeronian 67 ; Placodermi, Telychian-Wenlock) 68 . ...
Full-text available
The ecological context of early vertebrate evolution is envisaged as a long-term trend towards increasingly active food acquisition and enhanced locomotory capabilities culminating in the emergence of jawed vertebrates. However, support for this hypothesis has been anecdotal and drawn almost exclusively from the ecology of living taxa, despite knowledge of extinct phylogenetic intermediates that can inform our understanding of this formative episode. Here we analyse the evolution of swimming speed in early vertebrates based on caudal fin morphology using ancestral state reconstruction and evolutionary model fitting. We predict the lowest and highest ancestral swimming speeds in jawed vertebrates and microsquamous jawless vertebrates, respectively, and find complex patterns of swimming speed evolution with no support for a trend towards more active lifestyles in the lineage leading to jawed groups. Our results challenge the hypothesis of an escalation of Palaeozoic marine ecosystems and shed light into the factors that determined the disparate palaeobiogeographic patterns of microsquamous versus macrosquamous armoured Palaeozoic jawless vertebrates. Ultimately, our results offer a new enriched perspective on the ecological context that underpinned the assembly of vertebrate and gnathostome body plans, supporting a more complex scenario characterized by diverse evolutionary locomotory capabilities reflecting their equally diverse ecologies.
... Representatives of this group are characterized, among other traits, by the presence of bony spines in front of all paired and median fins except the caudal (Denison, 1979), which has given rise to their colloquial name of 'spiny sharks'. The occurrence of pectoral fin spines is recognized as a potential gnathostome synapomorphy (Miller et al., 2003) or symplesiomorphy (Coates, 2003), being also present in other major groups of Paleaozoic jawed vertebrates, including placoderms (Young, 2010), 'non-acanthodian' chondrichthyans (Miller et al., 2003), and osteichthyans (Zhu et al., 1999). However, this trait was independently lost in the later evolutionary history of these lineages and is absent in most living representatives (Coates, 2003;Miller et al., 2003), with the exception of catfishes (Siluriformes), that acquired pectoral fin spines as an evolutionary reversion (Price et al., 2015). ...
Full-text available
A recent analysis of the vascularization of the pectoral fin in antiarchs indicated that they resembled jawless osteostracans rather than other jawed vertebrates, thereby challenging the monophyly of the class Placodermi. Examination of the evidence proposed to support this new hypothesis shows misinterpretation of well-established morphology in a range of antiarchs, with incorrect homologies being applied to conclude that the subclavian artery and vein originated in the back of the branchial chamber. The interpretation is rejected and evidence is summarized showing that pectoral fin vascularization in antiarchs conforms with that of other jawed fishes. The position of the antiarchs as a major subgroup of the placoderm fishes is confirmed, with four antiarch characters (external endolymphatic openings; palatoquadrate-suborbital plate complex; Meckel's cartilage-infragnathal connection; extensive postorbital endocranial processes) providing additional support for placoderm monophyly.
Full-text available