ArticlePDF Available

Non-ammocoete larvae of Palaeozoic stem lampreys

Authors:

Abstract and Figures

Ammocoetes—the filter-feeding larvae of modern lampreys—have long influenced hypotheses of vertebrate ancestry1–7. The life history of modern lampreys, which develop from a superficially amphioxus-like ammocoete to a specialized predatory adult, appears to recapitulate widely accepted scenarios of vertebrate origin. However, no direct evidence has validated the evolutionary antiquity of ammocoetes, and their status as models of primitive vertebrate anatomy is uncertain. Here we report larval and juvenile forms of four stem lampreys from the Palaeozoic era (Hardistiella, Mayomyzon, Pipiscius, and Priscomyzon), including a hatchling-to-adult growth series of the genus Priscomyzon from Late Devonian Gondwana. Larvae of all four genera lack the defining traits of ammocoetes. They instead display features that are otherwise unique to adult modern lampreys, including prominent eyes, a cusped feeding apparatus, and posteriorly united branchial baskets. Notably, phylogenetic analyses find that these non-ammocoete larvae occur in at least three independent lineages of stem lamprey. This distribution strongly implies that ammocoetes are specializations of modern-lamprey life history rather than relics of vertebrate ancestry. These phylogenetic insights also suggest that the last common ancestor of hagfishes and lampreys was a macrophagous predator that did not have a filter-feeding larval phase. Thus, the armoured ‘ostracoderms’ that populate the cyclostome and gnathostome stems might serve as better proxies than living cyclostomes for the last common ancestor of all living vertebrates.
Adult and juvenile stages of P. riniensis a–g, AM 5750. h–n, AM 7538. a, b, Main slab of AM 5750. Photograph (a) with angled single-directional light (unlike the vertical polarized light in Fig. 1a); and photograph from Fig. 1a overlain with outlines of the anatomical structures identified in this study (b). For interpretive drawing, see Fig. 1b. c, d, Counterslab of AM 5750. Photograph (c) and interpretive drawing (d). e, Detailed, low-angle-light photograph of the oral funnel, showing the individual cusps of the circumoral feeding apparatus and ridges along the edge of the oral funnel. f, g, Detailed anatomy of the region surrounding the left (f) and right (g) eye, showing eye lenses and otic capsules. Owing to the thickness of the preserved body, some of the dorsal skeletal structures in the snout (for example, tectal cartilages) cannot be observed at the surface. h, i, Main slab of AM 7538. Photograph overlain with outlines of the anatomical structures identified in this study (h) and interpretive drawing (i). j, Detailed photograph of the branchial basket on main slab. k, Detailed photograph of the left eye, showing the periocular structure. l, Photograph of counterslab of AM 7538. m, n, Detailed low-angle-light photographs of the circumoral feeding apparatus, which consists of 14 grooved, petaliform plates. m, Main slab, showing cast. n, Counterslab, showing mould. Scale bar, 5 mm (a–d), 2 mm (h, i, l). Definitions of abbreviations are provided in the legends of Figs. 1–3 and ‘Anatomical abbreviations’ in the Methods.
… 
Late larval stages of P. riniensis a–f, AM 5816. g–k, AM 5815. l, m, AM 7539. a, b, Main slab of a large late larva, AM 5816 (no counterslab was found). Photograph overlain with outlines of the anatomical structures identified in this study (a) and interpretive drawing (b). c, Detailed photograph of the head, showing potential preservation of the petaliform plates of the circumoral feeding apparatus (arrowhead) and additional soft tissue structures. d, Detailed photograph of the branchial basket and digestive tract. e, f, Detailed photographs of the posterior trunk, with broken lines showing partial outline of the tail. g, h, Main slab of AM 5815, a small late larva missing a large portion of the snout and posterior half of the trunk. Photograph overlain with outlines of the anatomical structures identified in this study (g) and interpretive drawing (h). i, Photograph of counterslab, which shows posterior extremity of the trunk. j, Detailed photograph of the head, showing arrangement of the branchial arches and positions of the snout structures. k, Detailed photograph of the snout and eyes, showing the small snout structures and otic capsules. l, m, Main slab of AM 7539, a small late larva (counterslab has no anatomically informative trace). Photograph overlain with outlines of the anatomical structures identified in this study (l), and interpretive drawing (m). Scale bars, 2 mm. Definitions of abbreviations are provided in the legends of Figs. 1–3 and ‘Anatomical abbreviations’ in the Methods.
… 
This content is subject to copyright. Terms and conditions apply.
408 | Nature | Vol 591 | 18 March 2021
Article
Non-ammocoete larvae of Palaeozoic
stem lampreys
Tetsuto Miyashita1,2,3,4,7 ✉, Robert W. Gess5,7, Kristen Tietjen1,6 & Michael I. Coates1
Ammocoetes—the lter-feeding larvae of modern lampreys—have long inuenced
hypotheses of vertebrate ancestry1–7. The life history of modern lampreys,
whichdevelop from a supercially amphioxus-like ammocoete to a specialized
predatory adult, appears to recapitulate widely accepted scenarios of vertebrate
origin. However, no direct evidence has validated the evolutionary antiquity of
ammocoetes, and their status as models of primitivevertebrate anatomy is uncertain.
Here we report larval and juvenile forms of four stem lampreys from the Palaeozoic era
(Hardistiella, Mayomyzon, Pipiscius, and Priscomyzon), including a hatchling-to-adult
growth series of the genus Priscomyzon from Late Devonian Gondwana. Larvae of all
four genera lack the dening traits of ammocoetes. They instead display features that
are otherwise unique to adult modern lampreys, including prominent eyes, a cusped
feeding apparatus, and posteriorly united branchial baskets. Notably, phylogenetic
analyses nd that these non-ammocoete larvae occur in at least three independent
lineages of stem lamprey. This distribution strongly implies that ammocoetes are
specializations of modern-lamprey life history rather than relics of vertebrate ancestry.
These phylogenetic insights also suggest that the last common ancestor of hagshes
and lampreys was a macrophagous predator that did not have a lter-feeding larval
phase. Thus, the armoured ‘ostracoderms’ that populate the cyclostome and
gnathostome stems might serve as better proxies than living cyclostomes for the last
common ancestor of all living vertebrates.
In evolutionary narratives of vertebrate origin, ammocoetes remain
a convenient model—if not an analogue—of hypothetical ancestral
states, because few alternatives are offered by the controversial fossil
forms attributed to the vertebrate stem or the anatomically specialized
living chordates8. The overall resemblance between ammocoetes and
cephalochordates drove early comparative analyses to postulate an
ammocoete-like, sand-burrowing filter feeder at the base of vertebrate
phylogeny
2–5
, and even to propagate the idea that cephalochordates
are ‘degenerate’ vertebrates
1
. Circumstantial support for these sce-
narios has begun to erode in recent times. Cephalochordates have now
been removed to the earliest branch of the chordate crown9; hagfish,
which have now been shown to be firmly allied as the sister group to
lampreys
10
, show no trace of an ammocoete-like larval phase
11
; and puta-
tive stem vertebrates (Metaspriggina and myllokunmingiids) from the
Cambrian period are morphologically remote from ammocoete-like
conditions (such as fused branchial arches or primordial sensory cap-
sules)
12,13
. However, in the absence of direct contradictory evidence,
ammocoete-like hypothetical ancestors persist in modern hypotheses
of vertebrate origin
1418
. These models both predict and require a deep
evolutionary origin of ammocoete morphology.
Here we present an analysis of ontogenetically immature specimens
of four Palaeozoic stem lampreys to test whether an ammocoete-like
phase existed early in the lamprey lineage. To date, the taxon Mesomyzon
mengae from Early Cretaceous lake beds of China represents the earli-
est known record of the ontogenetic transition from ammocoete to
adult19. However, the stem taxa reported here are at least 180million
years older than Mesomyzon and all of them derive from non-freshwater
environments. These exceptionally well-preserved
20
specimens show
no distinctive ammocoete-like traits. Instead, these specimens exhibit
characteristics that are associated with the adult phase of modern lam-
preys. In each fossil taxon, the immature individuals signal a transfor-
mation series that does not parallel the ontogeny of modern lampreys.
Priscomyzon riniensis, which is from the latest Devonian stage (the
Fammenian) of Waterloo Farm (South Africa), provides the earliest
and most complete ontogenetic series among the four stem lampreys
that we examined. In the original description21, the holotype (acces-
sion code: Albany Museum (AM) 5750) is the only specimen reported,
and it remains the sole representative of the adult stage in our newly
reconstructed series (Fig.1a, b, Extended Data Fig.2a–g). AM5750 has
prominent eyes22 (which were originally interpreted as otic capsules21),
a well-developed oral funnel with an annular cartilage, a ring of 14cusps
in the circumoral feeding apparatus, and 7pairs of branchial arches.
Seven additional specimens of the same taxon are reported here for
the first time. These come from the same locality as the holotype and
represent younger ontogenetic stages (Fig.1d–q). The smallest of these
(AM5813, which has a body length of 14mm) carries an abdominal bulge
https://doi.org/10.1038/s41586-021-03305-9
Received: 14 January 2020
Accepted: 28 January 2021
Published online: 10 March 2021
Check for updates
1Department of Organismal Biology and Anatomy, University of Chicago, Chicago, IL, USA. 2Canadian Museum of Nature, Ottawa, Ontario, Canada. 3Department of Biology, University of
Ottawa, Ottawa, Ontario, Canada. 4Department of Earth Sciences, Carleton University, Ottawa, Ontario, Canada. 5Albany Museum and the Department of Geology, Rhodes University,
Makhanda, South Africa. 6Biodiversity Institute & Natural History Museum, University of Kansas, Lawrence, KS, USA. 7These authors contributed equally: Tetsuto Miyashita, Robert W. Gess.
e-mail: TMiyashita@nature.ca
Nature | Vol 591 | 18 March 2021 | 409
of low tissue density (Fig.1p, q) that extends from behind the branchial
basket to the anal region. We tentatively identify this structure as a yolk
sac, and therefore interpret AM5813 as a hatchling.
From hatchling to adult, all eight specimens of Priscomyzon (Fig.1a–q,
Extended Data Figs.1–5) exhibit prominent eyes and anteroposteriorly
short branchial baskets that are fused at the pericardial region. Tectal
cartilages are partially preserved in the four youngest specimens, and
a funnel with the circumoral feeding apparatus is present in all but one
specimen (AM5817), which is missing the snout. In the immature indi-
viduals (AM7538 and AM5813) the feeding apparatus consists of 14pet-
aliform units, whereas in the adult (AM5750) the apparatus is preserved
as a ring of 14cusps that resembles the circumoral keratinous teeth of
modern adult lampreys. All of these characters (a circumoral feeding
apparatus, prominent eyes, tectal cartilages, and posteriorly closed
branchial baskets) occur in the adult phase of modern lampreys, but
never prior to metamorphosis23. In ammocoetes, the sensory capsules
are underdeveloped (for example, ophthalmic development is arrested
in the form of eyespots); the upper lips form a hood rather than a funnel;
the branchial baskets are long with independent left and right sides
that are aligned in parallel; and tectal cartilages are absent (Fig.1r, s).
A unique sequence of morphological changes across ontogeny
is evident in the Priscomyzon series. The diameter of the oral funnel
appears to increase in positive allometry against body width, whereas
the elongate snout becomes progressively shorter. The branchial bas-
kets also increase in proportion as the branchial arches become radially
arranged. In the adult (AM5750), the first two branchial arches extend
anterior to the eye, whereas they are immediately posterior toor level
withthe eye in younger individuals (Fig.1h–k). Although the juvenile
specimen (AM7538) is similar to the adult in discrete morphological
characters, it is no larger than the larvae—and may even be shorter in
body length than one larva (AM5817). However, a composite size trait
(the first principal component of 13 metric traits) corroborates the
ac cfa
of
ps
iiiiii
iv
v
vi
vii
el
pc
tr
nhp
1
2
3
4
5
6
7
bb
pbl
oc
cbb
bl
dt
ys
nc?
cbb
oc
pbl
bb
nhp?
tr
cfa iii
iii
iv
v
vi
vii
tc
iii
iii
iv
v
vi
vii
cfa
el
bb
1
2
3
4
5
6
7pos
cbb
nhp
bb
cbb
cfa
ac?
nhp
tc
tr
pc
oc
cbb
1
2
3
4
pbl
5
6
7bb
dt
cfa
ac
tr
pc
viii
cbb
bb
dt
tc
cbb
bb
nc
tc
cfa
cbb
bb
nc
oh nhp
es oc nc
bb ht
dt
ys
ee
e
e
e
e
ee
ee
e
e
ee
eebb
oc
ys
AM 5750
AM 5813
AM 7538
AM 5816
AM 5815
AM 7539 AM 5817
AM 5814
ab
c
def
gi
h
jk
lm
no
pqr
s
Fig. 1 | A growt h series of Priscomyzon riniensis. Th is growth seri es has no
ammocoe te phase, and is sh own in reverse onto genetic order. a, b, Adult stage ,
represen ted by the holoty pe AM5750, shown as a pho tograph (a) and
interpret ive drawing (b). The an atomy is reinter preted from the or iginal
description21 (‘A-4. Additional d escription s’ section o ftheSupplementar y
Infor mati on). c, Schematic re construc tion of a part of th e circumoral feedi ng
apparatus of Priscomyzon, showing t wo petaliform pla tes and cusps f rom
oblique ventr al view from a cen tre of the oral funne l. d, e, Juvenile stage,
represen ted by AM7538. This s tage is charac terized by a well- developed oral
funnel and b ranchial regio n. fk, Late larval st age, represen ted by AM5816
(f, g), AM5815 (h, i), and AM7539 (j, k). This s tage is interm ediate in snout
length an d branchial expan sion. lo, Early lar val stage, repre sented by AM5817
(l, m) and AM5814 (n, o). The larvae are sle nder and elonga te, and the branch ial
region is sm all compared to the r est of the body. p, q, Hatchlin g stage,
represen ted by AM5813. T his 14-mm-long spe cimen is charac terized by an
abdominal b ulge (which we identif y as a potentia l yolk sac) andhas an elongat e
snout and sma ll branchial reg ion. Prominen t eyes and a circumora l feeding
apparatusare alre ady present in t his individual a s in later ontogen etic stages .
r, s, Early ontogenetic p hases of the mo dern lamprey Petromyzon marinus.
A hatchling ( Tahara’s stage 26) (r) and an early ammoco ete (Tahara’s stage 30)
(s) in left lateral v iew. Unlike in the stem lam prey, larvae of modern l ampreys
have an oral hood , primordial eye spot s, and elongate b ranchial basket s. Scale
bars, 2mm. ac, an nular cartila ge; bb, branchial basket ; bl, bi-lobed per icardial
struct ure; cbb, pericardi al commissure b etween branc hial baskets; cfa ,
circumoral fe eding apparatu s; dt, digestive tra ct; e, eye; el, eye lens; es, eye
spot; ht, hea rt; nc, notoch ord; nhp, nasohypoph seal organ; oc, o tic capsule; of,
oral funnel; oh , oral hood; pbl, pos tbranchial lat eral structu re; pc, parachordal
commissu re; pos, perioc ular structur e; ps, piston syste m; tc, tectal c artilage;
tr, trabecular c artilage; ys, yolk sa c. Arabic nume rals denote bran chial arches,
and Roman numerals indicate elements of the circumoral feeding apparatus.
410 | Nature | Vol 591 | 18 March 2021
Article
ontogenetic sequence that was reconstructed from discrete morpho-
logical characters, and places AM7538 between the adult and larval
specimens (Extended Data Fig.9b, c). This is consistent with modern
lamprey ontogeny, in which body size decreases across the larva–juve-
nile transition (the largest ammocoetes reach nearly 1.5times the body
length of the smallest adults)24.
Other Palaeozoic stem lampreys also provide evidence of
non-ammocoete larvae. We identified two specimens of Pipiscius
zangerli that carry a yolk sac (Fig.2a–d) from the Mazon Creek fauna
(dating to the Moscovian age from the Pennsylvanian subperiod of
Illinois). The smaller of the pair (accession code Royal Ontario Museum
(ROM) 56679) has a body length of 19mm and carries a proportionally
larger sac than that of FMNH (Field Museum of Natural History) PF16082
(which has a body length of 32mm). As with Priscomyzon, these hatch-
lings are unlike ammocoetes. Both ROM56679 and FMNHPF16082 have
prominent eyes, an oral funnel, and small branchial baskets relative to
body length. Despite their diminutive size and association with a yolk
sac, the character suites of these specimens are otherwise similar to
those of larger specimens of Pipiscius. Most diagnostically for this
taxon, these characters include a feeding apparatus that is shown here
to consist of concentric rings of petaliform plates and cusps (Fig.2e,
f, Extended Data Fig.6a–h), similar to (but more numerous and prob-
ably more complex than) that of Priscomyzon. These similarities sig-
nal the affinity of Pipiscius among stem lampreys, corroborating the
original taxonomic assignment
25
that has long lacked unambiguous
morphological support
10,26,27
. A corollary of this comparison is that
such plates and cusps probably co-occurred in the feeding apparatus
of Priscomyzon and are differentially exposed or preserved in individual
specimens (Fig.1c).
One larval specimen of the stem lamprey Mayomyzon pieckoensis28
(FMNH PF8167) is known from the same locality as Pipiscius. FMNH
PF8167 is approximately one third of the length of the largest known
specimen of Mayomyzon (Fig.3) and corresponds to the late larva in
the Priscomyzon series (Fig.1h, i). Consistent with the larvae of Prisco-
myzon, it has prominent eyes, tectal cartilages, and a well-developed
oral funnel. In comparison to other specimens of Mayomyzon
29
, the
branchial region is proportionally greater in the larva (Extended Data
Fig.6i–v). The Mayomyzon series also contains specimens from sev-
eral ontogenetic stages in which the gut imprint is independent from
the branchial region (Extended Data Fig.6i–v). This trait differs from
modern ammocoetes, in which the branchial region communicates
directly with the gut. In modern lampreys, the gut becomes separated
from the branchial passage only during metamorphosis, which enables
suctorial pumping in adult lampreys30. This ontogenetically terminal
state in modern lampreys is probably the general condition among
the Palaeozoic stem taxa, regardless of ontogenetic stage, because
its correlate—a dorsally positioned gut or posteriorly joined branchial
basket—is observed in all four taxa described in this Article (Extended
Data Figs.2–7).
Furthermore, we identified new specimens to supplement a previ-
ously described series of late ontogenetic stages
3133
in Hardistiella
montanensis, a stem lamprey from the Bear Gulch fauna (dating to the
Bashkirian age from the Pennsylvanian of Montana). The smallest speci-
men of Hardistiella (CM (Carnegie Museum of Natural History)4505)
(Extended Data Fig.7a, b) is approximately two thirds the estimated
length of the largest (CM63079 and UMPC (University of Montana
Paleontological Collections)10210), and has previously been described
as a late-stage larva
31
. Including the two new specimens (ROM78122 and
UMPC10210), the Hardistiella series shows decreasing body depth with
growth, which culminates in an anguilliform profile similar to that of
modern adult lampreys (Extended Data Figs.7a–l, 9f).
Ontogenetic transformations inferred from these Palaeozoic stem
lampreys point consistently to the absence of an ammocoete phase.
Although the Palaeozoic larval specimens are larger than modern
lamprey hatchlings (which are approximately 10mm in length), the
e
a
b
c
d
f
Groove
inlls
Cusp
inlls
e
e
e
e
tc?
of
cfa
dt
ys
an
ba1234ba
bb
oc?
bb
of
cfa
dt
an
ys
Fig. 2 | Hatc hlings of Pipis cius zangerli. The se specimen s show anatomical
traits com parable to thos e of Priscomyzon. ad, ROM56679 (a, b) and
FMNHPF16082 (c, d) have a yolk sa c and a circumoral fee ding apparatus t hat
consist s of petaliform un its. e, f, Three-dimensional micro-computed
tomograp hy renderings s how part and coun terpart of the c ircumoral feedin g
apparatus i n a mature specim en of Pipiscius (FMNHPF8344). Eac h petaliform
plate is split i nto lobes by a long itudinal groove towa rd the outer edge. At l east
three (and pote ntially four) rings of cu sps are preser ved (those demarc ating
the anteri or-most plate are highlig hted), and each cus p occurs at the pla te
boundari es. Scale bar s, 2mm. an, anus; ba, bran chial arch; other abb reviations
as in Fig.1.
a
b
1234567
bb
e
e
tc
ac? dt pcs?
oc
of tr
Fig. 3 | Lar val specime n of M.pieckoensis (FMNH PF8167). T his specimen
shows anatom ical traits co mparable to thos e of Priscomyzon: it has an elon gate
snout, promin ent eyes, and sep aration bet ween digestive t ract and branc hial
passage . Scale bar, 2mm. pcs, per icardial struc ture; other abbr eviations as in
Fig.1.
Nature | Vol 591 | 18 March 2021 | 411
former are sufficiently small as to make a preceding ammocoete phase
extremely unlikely. Otherwise, ammocoete traits would have to develop
early, only to metamorphose at diminutive size into the forms repre
-
sented by AM5813 and ROM56679 (Figs.1n, o, 2a, b). By contrast, living
lampreys persist as filter-feeding ammocoetes for 2–7years, grow
much larger than these Palaeozoic larvae, and usually metamorphose
at lengths of 150–250mm
24
. Although no hatchlings have been identi-
fied in Hardistiella and Mayomyzon, the similarities that are manifest
in the later ontogenetic stages add support to the generality of the
ontogenetic pattern thatwe observed in Priscomyzon (Extended Data
Fig.9). This reconstructed life history for the stem lampreys appears
to be consistent with palaeoecological settings. All three localities
represent protected, shallow, marine-influenced habitats in which
larvae, juveniles, and sexually mature and/or gravid adults of other
fishes (placoderms, chondrichthyans, actinopterygians, and sarcop-
terygians) occur
34–38
. It is difficult to determine whether these stem
lampreys represent direct or indirect developers
13
. In Priscomyzon,
immature and adult forms are set apart by several characters (Extended
Data Fig.9c), and the transition may therefore qualify as metamorpho-
sis. A possible reduction in body length across this transition is also
reminiscent of metamorphosis in modern lampreys
24
(Extended Data
Fig.9b). However, there is no clear evidence for the marked dietary
shift that is usually associated with metamorphosis.
With the addition of these Palaeozoic stem lampreys, the results of
our phylogenetic analysis support a reverse character polarity to the
prediction of ammocoete-based hypotheses of vertebrate origins:
that is, the life history of modern lampreys represents a derived con-
dition39,40. Life-history traits that are distinct from those of modern
lampreys—including large hatchling size relative to the adult, and the
lack of evidence for filter-feeding larvae—occur in three stem-lamprey
branches and successive outgroups of the lamprey clade(Fig.4). This
character distribution offers two further implications: (a) similar to
hagfish and modern adult lampreys, the last common ancestor of living
cyclostomes was probably a macrophagous predator; and (b) condi-
tions at the crown vertebrate node might be better informed by the
armoured taxa that occupy both the cyclostome and gnathostome
stems (namely, the ostracoderms) (Extended Data Fig.10). Thus, the
cyclostome and gnathostome crowns each represent an independ-
ent evolution of carnivory that emerged from the stem assemblage
of deposit feeders such as anaspids, galeaspids, osteostracans, and
thelodonts7.
Given this fossil evidence, modern ammocoetes do not appear to
provide a sound model for the hypothetical ancestral conditions of
vertebrates. Therefore, morphological similarities between ammo-
coetes and invertebrate chordates might be attributed to convergence
or reversal facilitated by similar life habits. For instance, an ammo-
coete has an endostyle—an infrapharyngeal organ that stores iodine
and secretes food-trapping mucous into the branchial system
40
—as
in invertebrate chordates. The ammocoete endostyle later develops
into a follicular thyroid, which is a vertebrate homologue of the chor-
date endostyle
41,42
. The derived state of an ammocoete is seemingly
at odds with this apparent recapitulation, but the ecophysiology of
modern lampreys offers a scenario that is compatible with both. In
the lamprey crown group, the prolonged filter-feeding larval phase
delays the requirement for the endocrine function of a thyroid in
iodine-poor freshwater environments40. This pre-metamorphic state
may have allowed the presumptive thyroid in common ancestors of
the crown-group lampreys to be co-opted as a functional equivalent
of those present in invertebrate chordates. Understanding whether
this co-option event represents convergence or reversal (and whether
it accompanied heterochronic shift) requires a detailed comparison
of the development and determination of the protochordate state of
the endostyle homologue.
Thus, lampreys and cephalochordates remain important branches
for constraining a suite of characters at the crown vertebrate node,
but neither is sufficient for the reconstruction of the ancestral state.
Paradoxically for a vertebrate clade with a notoriously poor-quality
fossil record8,26, here we have robust evidence of a substantial change
in life-history strategy, despite the conservation of adult morphology
of 360million years or more. An ammocoete phase (and the distinct
metamorphosis from filter feeders to predators) evolved somewhere
along the nearly 200-million-year ghost lineage (Fig.4) during which
lamprey habitats broadened to include freshwater. In a clear contrast
to the Palaeozoic stem lampreys
28,36,38
, the early ontogenetic stages of
post-Palaeozoic lampreys (whether fossil or living) occur exclusively in
freshwater environments13,19,24. This implied ecological shift suggests
that ammocoetes were an adaptation to closed, fluctuating fluvial
and lacustrine habitats in which filter feeders are typically less diverse
Haikouella
Myllokunmingia
Haikouichthys
Metaspriggina
Myxineidus
Hardistiella
Mayomyzon
Pipiscius
Priscomyzon
Mesomyzon
Geotria
Mordacia
Lampetra
Lethenteron
Petromyzon
Gilpichthys
Myxinikela
Tethymyxine
R. eos
R. lopheliae
E. burgeri
E. stoutii
E. atami
Myxine
Neomyxine
Euconodonta
Anaspida
Gnathostomata (total group)
Ammocoete-like larval stage
Gametes/hatchlings relatively large
?
?
?
?
?
+
+
+
+
+
+
?
?
?
?
?
?
?
?
?
?
?
?
+
?
+
+
?
+
?
?
?
?
+
+
+
+
?
?
?
/+
Vertebrata
(crown group)
Cyclostomi
(crown group)
Cyclostomi (total group)
Petromyzontiformes (total group)Myxinoidea (total group)
Date (Ma)
500 400 300 200 100 0
C O S D M Pn P T J K Pg N
Fig. 4 | Time -calibra ted phylogen etic tree of ea rly vertebrat e lineages
shows a late evolut ionary ori gin of a fil ter-feedin g larva. The t ree shows a
portio n of strict con sensus of 360mos t parsimonious t rees based o n 52taxa
and 167morphologi cal characte rs, revised f rom a previous stud y10; the full tree
is shown in Ex tended Data F ig.10. Mean divergenc e estimates for n odes are
also base d on the previous s tudy10. Tips (squares) show the m odern
lamprey-like (blue) an d non-modern-lam prey-like (red) state for the t raits that
are compared i n the right colum ns. Black indic ates missing i nformation. Ti ps
for composit e taxa (bars) show th e chronologic al range of the fossil re cord for
those clad es. Nodes (empt y circle, total no de; filled circ le, crown node) are als o
colour-coded, w ith the ances tral state rec onstruct ed parsimoniou sly and
unambigu ously. Icons asso ciated with ea ch lamprey taxon de note the state o f
larval phase (blue silhouette, ammocoete; red silhouette, non-ammocoete)
and whethe r a circumoral feedi ng apparatus th at consists of p etaliform plate s
and cusps is p resent (compute d tomography ren dering in red). We scor ed
relative size s of gametes or h atchlings in Pipiscius and Priscomyzon on the basis
of the smalle st specimen s describe d in this Articl e (Figs.1p, q, 2a, b), and in
Gilpichthys25 and Hardistiella32 on th e basis of the ser ies of abdomin al
struct ures that were previ ously interpre ted as gonads (Ex tended Dat a
Fig.7i–p). Anaspida wa s scored on the bas is of Euphanerops. Scale at
thebottom sh ows approximate date i n millions of years a go (Ma) along with
period or s ubperiod bo undaries; C, Ca mbrian; O, Ordovician; S, S ilurian;
D, Devonian; M, Mis sissippian; Pn, Pe nnsylvanian; P, Permian; T, Triassic;
J, Jurassic ; K, Cretaceou s; Pg, Palaeogen e; N, Neogene.In th e hagfish c rown
group: E., Eptatretus; R., Rubicundus. E. atami is also known as a sp ecies of the
genus Paramyxine.
412 | Nature | Vol 591 | 18 March 2021
Article
than in marine communities. The presence of such a generalist larval
phase prior to the specialized adult form may have contributed to the
long-term survival of this lineage, after the apparent extinction of their
exclusively non-freshwater relatives.
Online content
Any methods, additional references, Nature Research reporting sum-
maries, source data, extended data, supplementary information,
acknowledgements, peer review information; details of author contri-
butions and competing interests; and statements of data and code avail-
ability are available at https://doi.org/10.1038/s41586-021-03305-9.
1. Haeckel, E. The Evolution of Man: A Popular Exposition of the Principal Points of Human
Ontogeny and Phylogeny3rd edn (Werner, 1876).
2. Gaskell, W. H. On the Origin of Vertebrates (Longmans, Green and Co., 1908).
3. Goodrich, E. S. Studies on the Structure and Development of Vertebrates (Dover, 1930).
4. De Beer, G. The Development of the Vertebrate Skull (Clarendon, 1937).
5. Romer, A. S. & Parsons, T. S. The Vertebrate Body5th edn (Saunders, 1977).
6. Gee, H. Before the Backbone: Views on the Origin of the Vertebrates (Springer, 1996).
7. Janvier, P. Early Vertebrates (Clarendons, 1996).
8. Janvier, P. in Major Transitions in Vertebrate Evolution (eds. Anderson, J. S. & Sues, H.-D.)
57–121 (Indiana Univ. Press, 2007).
9. Delsuc, F., Brinkmann, H., Chourrout, D. & Philippe, H. Tunicates and not cephalochordates
are the closest living relatives of vertebrates. Nature 439, 965–968 (2006).
10. Miyashita, T. etal. Hagish from the Cretaceous Tethys Sea and a reconciliation of the
morphological–molecular conlict in early vertebrate phylogeny. Proc. Natl Acad. Sci.
USA 116, 2146–2151 (2019).
11. Oisi, Y., Ota, K. G., Kuraku, S., Fujimoto, S. & Kuratani, S. Craniofacial development of
hagishes and the evolution of vertebrates. Nature 493, 175–180 (2013).
12. Janvier, P. Facts and fancies about early fossil chordates and vertebrates. Nature 520,
483–489 (2015).
13. Evans, T. M., Janvier, P. & Docker, M. F. The evolution of lamprey (Petromyzontida) life
history and the origin of metamorphosis. Rev. Fish Biol. Fish. 28, 825–838 (2018).
14. Gans, C. in The Skull. Volume 2. Patterns of Structural and Systematic Diversity (eds.
Hanken, J. & Hall, B. K.) 1–35 (The Univ. of Chicago Press, 1993).
15. Mallatt, J. Ventilation and the origin of jawed vertebrates: a new mouth. Zool. J. Linn. Soc.
117, 329–404 (1996).
16. Northcutt, R. G. The new head hypothesis revisited. J. Exp. Zool. B Mol. Dev. Evol. 304B,
274–297 (2005).
17. Cattell, M., Lai, S., Cerny, R. & Medeiros, D. M. A new mechanistic scenario for the origin
and evolution of vertebrate cartilage. PLoS ONE 6, e22474 (2011).
18. Jandzik, D. etal. Evolution of the new vertebrate head by co-option of an ancient chordate
skeletal tissue. Nature 518, 534–537 (2015).
19. Chang, M. M., Wu, F., Miao, D. & Zhang, J. Discovery of fossil lamprey larva from the Lower
Cretaceous reveals its three-phased life cycle. Proc. Natl Acad. Sci. USA 111, 15486–15490
(2014).
20. Sansom, R. S., Gabbott, S. E. & Purnell, M. A. Atlas of vertebrate decay: a visual and
taphonomic guide to fossil interpretation. Palaeontology 56, 457–474 (2013).
21. Gess, R. W., Coates, M. I. & Rubidge, B. S. A lamprey from the Devonian period of South
Africa. Nature 443, 981–984 (2006).
22. Sansom, R. S., Freedman, K., Gabbott, S. E., Aldridge, R. J. & Purnell, M. A. Taphonomy
and afinity of an enigmatic Silurian vertebrate, Jamoytius kerwoodi White. Palaeontology
53, 1393–1409 (2010).
23. Hardisty, M. W. in The Biology of LampreysVol. 3 (eds. Hardisty, M. W. & Potter, I. C.)
118–124 (Academic, 1981).
24. Renaud, C. B. Lampreys of the World. An annotated and Illustrated Catalogue of Lamprey
Species Known to Date (Food and Agriculture Organization of the United Nations, 2011).
25. Bardack, D. & Richardson, E. S. New agnathous ishes from the Pennsylvanian of Illinois.
Fieldiana Geol. 33, 489–510 (1977).
26. Janvier, P. Early jawless vertebrates and cyclostome origins. Zoolog. Sci. 25, 1045–1056
(2008).
27. Gabbott, S. E. etal. Pigmented anatomy in Carboniferous cyclostomes and the evolution
of the vertebrate eye. Proc. R. Soc. Lond. B 283, 20161151 (2016).
28. Bardack, D. & Zangerl, R. in The Biology of LampreysVol. 1 (eds. Hardisty, M. W. & Potter, I. C.)
1–65 (Academic, 1971).
29. Bardack, D. & Zangerl, R. First fossil lamprey: a record from the Pennsylvanian of Illinois.
Science 162, 1265–1267 (1968).
30. Hardisty, M. W. Biology of the Cyclostomes (Springer, 1979).
31. Lund, R. & Janvier, P. A second lamprey from the Lower Carboniferous (Namurian) of Bear
Gulch, Montana (U.S.A.). Geobios 19, 647–652 (1986).
32. Janvier, P. & Lund, R. Hardistiella montanensis n. gen. et sp. (Petromyzontida) from the
Lower Carboniferous of Montana, with remarks on the afinities of the lampreys. J.
Vertebr. Paleontol. 2, 407–413 (1983).
33. Janvier, P., Lund, R. & Grogan, E. D. Further consideration of the earliest known lamprey,
Hardistiella montanensis Janvier and Lund, 1983, from the Carboniferous of Bear Gulch,
Montana, U.S.A. J. Vertebr. Paleontol. 24, 742–743 (2004).
34. Lund, R. Chondrichthyan life history styles as revealed by the 320 million years old
Mississippian of Montana. Environ. Biol. Fishes 27, 1–19 (1990).
35. Grogan, E. D. & Lund, R. Soft tissue pigments of the Upper Mississippian
chondrenchelyid, Harpagofututor volsellorhinus (Chondrichthyes, Holocephali) from the
Bear Gulch Limestone, Montana, USA. J. Paleontol. 71, 337–342 (1997).
36. Lund, R., Greenfest-Allen, E. & Grogan, E. D. Habitat and diversity of the Bear Gulch ish:
life in a 318 million year old marine Mississippian bay. Palaeogeogr. Palaeoclimatol.
Palaeoecol. 342–343, 1–16 (2012).
37. Sallan, L. C. & Coates, M. I. The long-rostrumed elasmobranch Bandringa Zangerl, 1969,
and taphonomy within a Carboniferous shark nursery. J. Vertebr. Paleontol. 34, 22–33
(2014).
38. Gess, R. W. & Whitield, A. K. Estuarine ish and tetrapod evolution: insights from a Late
Devonian (Famennian) Gondwanan estuarine lake and a southern African Holocene
equivalent. Biol. Rev. 95, 865–888 (2020).
39. Hardisty, M. W. in The Biology of LampreysVol. 4b (eds. Hardisty, M. W. & Potter, I. C.)
165–259 (Academic, 1982).
40. Youson, J. H. & Sower, S. A. Theory on the evolutionary history of lamprey
metamorphosis: role of reproductive and thyroid axes. Comp. Biochem. Physiol. B
Biochem. Mol. Biol. 129, 337–345 (2001).
41. Ogasawara, M., Di Lauro, R. & Satoh, N. Ascidian homologs of mammalian thyroid
peroxidase genes are expressed in the thyroid-equivalent region of the endostyle. J. Exp.
Zool. 285, 158–169 (1999).
42. Ogasawara, M., Shigetani, Y., Suzuki, S., Kuratani, S. & Satoh, N. Expression of Thyroid
transcription factor-1 (TTF-1) gene in the ventral forebrain and endostyle of the agnathan
vertebrate, Lampetra japonica. Genesis 30, 51–58 (2001).
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in
published maps and institutional afiliations.
© The Author(s), under exclusive licence to Springer Nature Limited 2021
Methods
No statistical methods were used to predetermine sample size. The
experiments were not randomized, and investigators were not blinded
to allocation during experiments and outcome assessment.
Anatomical abbreviations
In addition to abbreviations used in Figs.1–3, Extended Data Figs.1–7
use the following abbreviations: af, anal fin; blb, longitudinal bar of
branchial basket; cf, caudal fin; df, dorsal fin; edg, broken edge; gn,
gonads; kd, kidney; kl, kaolinite lump; mo, mouth; my, myomere; ofr,
ridges on oral funnel; os, oral structure; ot, otic capsule; and plp, pos-
terior lateral plate.
Provenance
All specimens of Priscomyzon described in this Article were excavated
and identified by R.W.G., from shale rescued from the roadworks at
Waterloo Farm near Makhanda (also known as Grahamstown) (South
Africa)
43
. All but the smallest specimen (AM5813) were collected from
the same lens as the type specimen (AM5750). AM5813 comes from
an immediately adjacent lens. The specimens are accessioned at the
Albany Museum. The specimens of the Mazon Creek stem lampreys
(Mayomyzon and Pipiscius) have previously been reported from private
collections
25,28
(and transferred to public repositories subsequently) or
were identified by T.M. and M.I.C. in the designated depositories (FMNH
and ROM). The specimens of Hardistiella were previously described3133,
or reported to T.M. by K. Seymour and D.C. Evans (ROM) and J. Pardo
(University of Calgary).
Anatomical reconstructions
We examined all specimens under a binocular microscope with chang-
ing light angles for the maximum refraction and topographical infor-
mation. In Priscomyzon, the fragility of the specimens precluded any
potentially corrosive methods such as alcohol immersion or ammonium
chloride coating. Their taphonomy includes partial metamorphosis of
the enclosing mudstone during deep burial, during which the tissue
remnants were completely replaced by secondary metamorphic mica,
which converted to kaolinite during uplift. This complete replacement
of the preserved tissues by kaolinite ruled out geochemical analysis
using radiation or spectrometric techniques. The Hardistiella series
comprises three specimens that have previously been described31–33,
and two previously undescribed specimens. Aside from the descrip-
tion of features relevant to the ontogeny of Hardistiella, a complete
description and detailed morphological analyses of the new specimens
are deferred at this time.
Taxonomic identification
Taxonomic identifications are based on several lines of inferences: char-
acters diagnostic to the genus (for example, the numbers of branchial
arches or circumoral elements) and specific to lampreys (for example,
oral funnel); faunal correlates (whether or not any other similar jawless
vertebrates occur in the locality); and fit to the previously known speci-
mens (whether or not the specimen falls into the range of morphological
and taphonomic variations known for the taxon). We assessed each
specimen against modern decay experiments20,4446 to consider decay
effects on preservation of the soft tissue characters (Extended Data
Fig.8). We then mapped the preserved characters across the specimen
series within each taxon to test for gradual morphological transitions
indicative of ontogenetic sequence, and assigned the specimens to a par-
ticular ontogenetic stage (Extended Data Fig.9). Detailed descriptions
of this ontogenetic analysis are available inSupplementary Information.
Computed tomography scan
The circumoral feeding apparatus of a mature individual of Pipis-
cius (FMNH PF8344) was scanned by K.T. at the Paleo CT Scan
Facility (Department of Organismal Biology and Anatomy, University of
Chicago). Scan parameters used for the part were: voxel size = 11.458µm;
voltage = 65kV; current = 175µA; timing = 2,000ms; filter = 0.5mm Cu.
Scan parameters used for the counterpart were: voxel size = 11.309µm;
voltage = 60kV; current = 180µA; timing = 1,000ms. No distinct layer
of soft tissue was discernible on part or counterpart but the parts did
not fit one another perfectly, which suggests that the body structures
were preserved in thin film (<50µm) at the part–counterpart interface.
Using Mimics (Materialise), K.T. generated an infilling layer over the
surface topography in both part and counterpart (Fig.2e, f). These
negatives show: (a) a mould of the cusps and cast of the petaliform
plates (Fig.2e); and (b) infillings for pulp cavities of the cusps and the
grooves between and within the plates (Fig.2f)
Illustrations
All specimens were illustrated with solid lines to represent topographi-
cal contours and/or changes in mineral compositions, grey shades to
represent organic imprints or tissues, and stipples to show anatomical
structures demarcated by topographical contours.All anatomical
drawings were prepared by T.M.
Phylogenetic analysis
We examined the relationships of the four stem lampreys in a dataset
modified from a previous analysis10. This dataset contains 52taxa and
167morphological characters, and was analysed using PAUP47. Strict
consensus of the 360most parsimonious trees (tree length=344; con-
sistency index=0.544; retention index=0.814; rescaled consistency
index=0.442) is shown in Extended Data Fig.10. Full analytical details
and dataset are available inSupplementary Information.
Reporting summary
Further information on research design is available in theNature
Research Reporting Summary linked to this paper.
Data availability
The digitally reconstructed part and counterpart of Pipiscius (Fig.2e,
f) remain under the copyright of the FMNH, and are available at https://
doi.org/10.6084/m9.figshare.13378628.v1. The original scan data are
property of the FMNH (depository: https://emudata.fieldmuseum.org/)
and are available upon request to the FMNH and the corresponding
author. The cladistic dataset used for our analysis is available asSup-
plementary Information.
43. Gess, R. W. High Latitude Gondwanan Famennian Biodiversity Patterns: Evidence from the
South African Witpoort Formation (Cape Supergroup, Witteberg Group) (Univ. of
Witwatersrand, 2011).
44. Sansom, R. S., Gabbott, S. E. & Purnell, M. A. Decay of vertebrate characters in hagish
and lamprey (Cyclostomata) and the implications for the vertebrate fossil record. Proc. R.
Soc. Lond. B 278, 1150–1157 (2011).
45. Sansom, R. S., Gabbott, S. E. & Purnell, M. A. Non-random decay of chordate characters
causes bias in fossil interpretation. Nature 463, 797–800 (2010).
46. Purnell, M. A. etal. Experimental analysis of soft-tissue fossilization: opening the black
box. Palaeontology 61, 317–323 (2018).
47. Swofford, D. L. PAUP* . (Sinauer, 2017).
48. State of New York Conservation Department. A Biological Survey of the Oswego River
System. Supplemental to Seventeenth Annual Report. (J. B. Lyon, 1928).
Acknowledgements We thank J.-B. Caron, J. Hanken, J. Hurum, P. Janvier, G. Clement,
Z. Johanson, O. Matton, K. Moore, T. Mörss, M. Purnell, S. Gabbott, T. Schossleitner, K. Seymour,
W. Simpson, and S. Walsh for collections access; M. Bronner, A. Chinsammy-Turan, S. Gess,
S. Green, D. Hockman, K. Miyashita, H. Parker, and R. Prevec for logistical support; J. Pardo for
reporting UMPC10210; K. Seymour and D. Evans for reporting ROM78122; and R. Plotnick for
the access to the D. Bardack slide collections. Funding was provided by Chicago Fellows
Program, Vanier Canada Graduate Scholarship, I. W. Killam Memorial Scholarship,
Commonwealth Science Conference Follow-on Grant (T.M.); NSF DEB-1541491 (M.I.C.);
Millennium Trust, South African DSI-NRF Centre of Excellence in Palaeosciences, National
Research Foundation, and the South African Natural Science Collections Facility (R.W.G.).
Author contributions T.M. designed the study, performed morphological and phylogenetic
analyses, prepared igures, and wrote the manuscript; R.W.G. conceptualized the study of
Article
Priscomyzon, performed the ieldwork in the Waterloo Farm lagerstätte, discovered the
specimens of Priscomyzon, and contributed to the morphological analysis of Priscomyzon and
drafting the manuscript; K.T. performed the computed tomography scan of Pipiscius and
provided reconstructions; M.I.C. coordinated the study, identiied the specimens of Pipiscius,
and contributed to drafting the manuscript.
Competing interests The authors declare no competing interests.
Additional information
Supplementary information The online version contains supplementary material available at
https://doi.org/10.1038/s41586-021-03305-9.
Correspondence and requests for materials should be addressed to T.M.
Peer review information Nature thanks Philippe Janvier, Shigeru Kuratani and the other,
anonymous, reviewer(s) for their contribution to the peer review of this work.
Reprints and permissions information is available at http://www.nature.com/reprints.
Extende d Data Fig. 1 | Com parison of th e specime ns of P.riniensis to the
same sca le in reverse onto genetic ord er. a, b, An adult (AM5750),
photograph (a) and interpre tive drawing (b). c, d, A juvenile (A M7538),
photograph (c) and interpret ive drawing (d). e, f, A large late larva (A M5816),
photograph (e) and interpret ive drawing (f). g, h, A small late lar va (AM5815),
photograph (g) and int erpretive drawi ng (h). i, j, A small late larva (A M7539),
photograph (i) and inter pretive drawing ( j). k, l, A large early lar va (AM5817),
photograph (k) and inter pretive drawing ( l). m, n, A small early larva (A M5814),
photograph (m) and interpre tive drawing (n). o, p, A hatchling (AM58 13),
photograph (o) and interpret ive drawing (p). Scale bar, 2mm.
q, Reconstr uction of thre e individuals of Priscomyzon, each repre senting a
different o ntogenetic st age. Clockw ise from the ri ght: a hatchlin g carrying a
yolk sac (bas ed on AM5813), tucked in th e meadow of the char ophyte
Octochara c rassa; a juveni le (based on A M7538), attached to the sub strate in
the foregrou nd; and an adult (bas ed on AM5750), looming over the ot her
individua ls and showing it s feeding appara tus. In the back ground, a schoo l of
the coelacanth Selenichthys kowiensis swi m above the charophy te meadow.
Artwor k by K.T.
Article
Extende d Data Fig. 2 | Se e next page for capti on.
Extende d Data Fig. 2 | Adul t and juvenile st ages of P.riniensis. ag, AM5750.
hn, AM7538. a, b, Main slab of A M5750. Photograph (a) with angled single-
directio nal light (unlike the ver tical polariz ed light in Fig.1a); and photogr aph
from Fig.1a overlain wit h outlines of the a natomical str uctures iden tified in
this study (b). For interpretive dr awing, see Fig .1b. c, d, Countersl ab of
AM5750. Photogr aph (c) and interpretive d rawing (d). e, Detailed, low-angle-
light photo graph of the oral f unnel, showing th e individual cus ps of the
circumoral fe eding apparatu s and ridges alon g the edge of the oral f unnel.
f, g, Detailed ana tomy of the region s urrounding the l eft (f) and right ( g) eye,
showing eye len ses and otic ca psules. Ow ing to the thickn ess of the pres erved
body, some of the do rsal skeletal st ructures in t he snout (for example, te ctal
cartila ges) cannot be obs erved at the sur face. h, i, Main slab of A M7538.
Photograp h overlain with outli nes of the anatom ical struct ures identif ied in
this study (h) and interpre tive drawing (i). j, Det ailed photogra ph of the
branchial ba sket on main slab. k, De tailed photog raph of the left e ye, showing
the perio cular struct ure. l, Photogra ph of countersla b of AM7538.
m, n, Detailed low-angle-light photographs of the circumoral feeding
apparatus , which consist s of 14grooved, petalifor m plates. m, Mai n slab,
showing ca st. n, Counter slab, showing mould . Scale bar, 5mm (ad), 2mm
(h, i, l). Definition s of abbreviatio ns are provided in th e legends of Figs .1–3 and
‘Anatomical ab breviations’ i nthe Methods.
Article
Extende d Data Fig. 3 | La te larval sta ges of P.riniensis. af, AM5816.
gk, AM5815. l, m, AM7539. a, b, Main slab of a large late la rva, AM5816
(no counterslab was found). Photograph overlain with outlines of the anatomical
struct ures identif ied in this st udy (a) and interpreti ve drawing (b). c, Detailed
photogra ph of the head, show ing potenti al preserva tion of the pet aliform plates
of the circumo ral feeding appa ratus (arrowhead) a nd additional s oft tissue
struct ures. d, Detailed pho tograph of the br anchial basket a nd digestive tra ct.
e, f, Detailed photo graphs of the p osterior tr unk, with broken li nes showing
partial ou tline of the tail . g, h, Main slab of AM5815, a sm all late larva mi ssing a
large port ion of the snout an d posterior h alf of the trunk . Photograph overla in
with outlin es of the anatom ical struc tures identi fied in this s tudy (g) and
interpret ive drawing (h). i, Photograph of counterslab, which shows posterior
extremit y of the trunk. j, D etailed pho tograph of the h ead, showing ar rangement
of the branch ial arches and po sitions of the s nout struct ures.
k, Detaile d photograph o f the snout and eyes , showing the smal l snout struc tures
and otic cap sules. l, m, Main slab of A M7539, a small late larva (countersl ab has no
anatomic ally informative tr ace). Photograph overla in with outline s of the
anatomic al structure s identif ied in this stud y (l), and interpreti ve drawing (m).
Scale bars , 2mm. Defini tions of abbrev iations are provi ded in the legen ds of
Figs.1–3 and ‘Anatomical abbrevi ations’ in theMe thods.
Extende d Data Fig. 4 | Ea rly larval sta ges of P.riniensis. ac, AM5817.
dj, AM5814. a, b, Main slab of AM5817, a large early l arva (no counters lab was
found). Photog raph overlain with ou tlines of the anat omical struc tures
identif ied in this stud y (a) and interpretive dr awing (b). c, Detailed photograph
of the head re gion, showing i ndividual bran chial arches and su rface erosion
struct ures. d, e, Main slab of AM5814, a small ea rly larva. Photo graph overlain
with outlin es of the anatom ical struct ures identif ied in this stud y (d) and
interpret ive drawing (e). f, g, Counterslab. Photograph (f) and interpret ive
drawing ( g), showing individu al branchial arche s. h, i, Detailed photographs of
the oral reg ion in main slab, with ( h) or without (i) broken li nes delineat ing the
circumoral feeding apparatus. j, Detailed ph otographs of th e head region in
countersl ab. Scale bars, 2mm. D efinitio ns of abbreviat ions are provided in t he
legends of Fi gs.1–3 and ‘Anatomical abbrev iations’ in theMet hods.
Article
Extende d Data Fig. 5 | Hat chling sta ge of P.rinien sis, re presente d by
AM5813 car rying a yolk sac . ac, Main slab. Photog raph (a), photograph
overlain with ou tlines of the ana tomical struc tures identi fied in this s tudy (b)
and interp retive drawing (c). d, e, Detaile d photographs o f the head regio n in
different light angles, highlighting different structures preserved by kaolinite
mass. f, Det ailed photogra ph of the branchial re gion. g, Detailed photograph
of the circumoral feeding apparatus. hj, Counterslab. Photograph (h),
interpret ive drawing (i) and de tailed photog raph of the head re gion ( j). Scale
bars, 2mm. Def initions of a bbreviation s are provided in the le gends of Figs.1–3
and ‘Anatomical a bbreviation s’ in theMethods .
Extende d Data Fig. 6 | Se e next page for capti on.
Article
Extende d Data Fig. 6 | Vario us growth sta ges of the Mazon C reek stem
lampreys P.zangerli and M.pieckoensis. av, Pipiscius (ah) and Mayomyzon
(iv) show suites of ch aracters tha t are compatible w ith the Priscomyzon seri es,
including pro minent eyes, an ora l funnel, and shor t branchial bas kets.
a, b, ROM56679, a hatchling carry ing a yolk sac. Phot ograph overlain wi th
outlines of t he anatomical s tructures i dentifie d in this study (a) and
photograph of the counterpart (b). c, d, FMNH PF16082, a hatch ling carry ing a
yolk sac. Phot ograph overlain wi th outlines of th e anatomical st ructures
identif ied in this stud y (c) and photograph of the c ounterpar t (d).
eh, Comparis on of the specim ens of P.zangerli at t he same scale re veals nearly
identica l suites of morph ological char acters bet ween the hatch ling and adult,
except size and p resence or abs ence of a yolk sac . e, f, Holotype FMNH PF8 346
shown as a phot ograph (e) and interp retive drawing (f), repres enting the
general adul t morpholog y of the taxon and pre served in comp arable
orienta tions to the two h atchling spec imens. g, The la rgest specim en known
for Pipiscius (FMNH PF83 44), showing the circumo ral feeding appar atus in
dorsal vie w. h, The smaller hatch ling (ROM56679), shown in interpre tive
drawing at th e same scale as eg (two a dult specime ns). ip, A specimen
referred to M.pieckoensis (FMNH PF8 167), representing a lar va. il, Main part .
Photograp h (i), photograph overlain w ith outlines of t he anatomical s tructures
identif ied in this stud y (j), inter pretive drawing ( k), and scanned phot ograph
taken in the 1960 s by D.Bardack to show the ori ginal state of t issue
preser vation in this spe cimen (l). The pho tograph in l was di scovered in the
slide collec tions of D.Bardack . A full view of the sno ut can be seen i n figure 2 of
ref. 28. The sp ecimen was da maged in the snout, a nd the thin film o f organic
tissues h as deteriora ted across the e ntire specime n. mp, Detailed
photogra phs and illustrat ion of the head re gion. Photog raph in low-angle
lighting to r eveal surface tex tures (m), interpretive dr awing (n), photograph in
high-angle li ghting to reveal th e film of prese rved tissue s (o) and photograph
of counterpart in high-angle lighting for comparison (p). q, Scan of photogra ph
taken in the 1960 s by D.Bardack, to show orig inal state of ti ssue preser vation in
the head re gion of FMNH PF568 7, a juvenile of M.pieckoensis. The pho tograph
in q was discovered i n the slide collec tion of D. Bardack . Further detail s can be
seen in f igures 2–4 of ref. 28. rv, Comparison of t he specimen s of M.pieckoensis
at the same s cale reveals chara cter transit ions across on togeny, including the
decreasi ng relative propor tions of branc hial region. r, s, Holoty pe FMNH
PF5687 shown a s a photograph (r) and int erpretive drawi ng (s), representing
the juvenile st age. t, Interp retive drawing of FM NH PF8167 at the same sc ale as
r, s (juvenile), u, v (adult). u, v, The largest know n specimen (RO M56787) shown
as a photogr aph (u) and interpreti ve drawing (v), representi ng the adult sta ge.
Scale bars , 2mm (ad, il, mp), 5mm (eh, rv). Definiti ons of abbreviat ions
are provided in t he legends of Fig s.1–3 and ‘Anatomic al abbreviatio ns’ inthe
Methods.
Extende d Data Fig. 7 | Se e next page for capti on.
Article
Extende d Data Fig. 7 | Grow th series o f H.montanensis. This seri es shows
suites of ch aracters tha t are compatible w ith those of othe r Palaeozoic ste m
lampreys, incl uding prominent e yes, an oral funnel, a nd short branc hial
baskets. T his series a lso document s gradual chara cter transit ions in the
decreasi ng body depth, in creasing inte rocular dista nce and progres sively
slender prof ile of the oral fu nnel. Compar ison reveals size di screpancie s across
the juvenile–ad ult transitio n and between t he sexually mature , gravid
specimen and other adults (discussed in section A-3a of the Supplementary
Informatio n in compariso n to modern lamprey s). al, Hardistiella
montanensis. a, b, Photograph (a) and i nterpretive draw ing (b) of late larva l
stage repre sented by CM4505, sho wing the deep er body than in any la ter
ontogenetic stages. c, d, Juvenile stage, repr esented by ROM781 22, showing
intermediate body depth. Photograph (c) and interpretive draw ing (d).
el, Adult stage re presented by CM63 079 (e, f), UMPC10210 (g, h) and
UMPC7696 (il). e, f, CM63079—incomplete, b ut potentially t he largest
specime n—shown as a photog raph (e) and interpret ive drawing (f).
g, h, UMPC10210, showing a n advanced sta ge of decay relative to o ther
specime ns, shown as a photo graph (g) and in terpretive draw ing (h).
il, UMPC7696 associa ted with pote ntial gonads (prob ably represen ting a
gravid femal e with ovaries or e ggs), shown as a photog raph (i) and inter pretive
drawing (l). k, De tailed photog raph of the abdo minal region of th e main part,
showing the c ircular struc tures interpre ted in ref. 32 as gonads . l, Tracing of the
anatomic al structure s in k (in grey). mp, The abdomina l structure s
wereidentif ied as gonads in t he stem myxino id Gilpichthysgreenei from th e
Mazon Creek fa una, indicati ng a small number of la rge ovaries or eg gs (white
arrowhead s) in this taxon. The se examples, wh ich are shown here in v arious
forms of develop ment and stat es of preserv ation, bolste r the gonadal
interpret ation for the ser ial, circular abdo minal struct ures in the holot ype of
the stem lamp rey Hardistiella UMPC7696 (k, l). In Gilpichthys, the stru ctures
appear to deve lop in a small number (3–5) of i ron concretion s as in ROM56389
(m), or paired serie s of fewer than a dozen circu lar depression s as in FMNH
PF234 64 (n). o, p, FMNH PF8464 shows a s eemingly ma ture state of
developmen t: 11circular stru ctures of iron-ri ch recryst allization, imp lying that
the conten ts were encaps ulated and chem ically distinc t from the rest of t he
body. The numb er agrees wi th the 11 prese nt in FMNH PF23 464, shown he re as a
photogra ph with low-angle lig hting (o) and as a detail ed photograph o f the
gonads in hig h-angle, high-ex posure lightin g (p). Scale bars, 5mm. Def initions
of abbreviat ions are provide d in the legends of Fi gs.1–3 and ‘Anatomical
abbreviat ions’ in theMetho ds.
Extende d Data Fig. 8 | Se e next page for capti on.
Article
Extende d Data Fig. 8 | Com parison of a natomical t raits prese rved in the
specim ens of Palaeo zoic stem lamp rey describ ed in this Ar ticle. a, Moder n
decay ser ies of ammoco etes and adult lam preys, after a prev ious publicat ion20.
Codes for the se reference de cay series are a s follows. Yellow, pristine; oran ge,
decayin g; red, onset of lo ss; terminal po int, complete lo ss. The x axis s hows
number of days , whereas yaxis ran ks decay sta ges. For each fossi l specimen
(bp), bars shown in orig inal colours (ye llow, orange, and red) repr esent
charact ers preserve d in the specime n; 50% transparen t (brown) bars repre sent
charact ers preserve d in the specime n but in a different for m than in the
modern refe rence; grey bars re present chara cters missi ng in the specime n; no
bars shown de notes that, in pr inciple, chara cters canno t be assesse d in the
specime n (for example, ‘slanting g ill openings’ c annot be deter mined in
dorsoventr ally compresse d specimens , whereas ‘gill sy mmetry ’ cannot be
assess ed in transvers ally compress ed specimen s). bi, Priscomyzon riniensis.
b, AM5750, represen ting the adult st age. c, AM7538, repres enting the juve nile
stage. df, AM5816 (d), AM5815 (e), and AM7539 (f), all representing th e late
larval stage. g, h, AM5817 ( g) and AM5814 (h), represen ting the early lar val
stage. i, AM5813, repres enting the hatc hling stage. j, k, Pipisc ius zangerli. ROM
5667 9 ( j) and FMNH PF160 82 (k), both represe nting the hatc hling stage.
l, Mayomyzon pieckoensis. FMNH PF8167, represen ting the late lar val stage.
mq, Hardistiella montanensis. m, CM4505, repr esenting the l ate larval sta ge.
n, ROM78122 , representin g the juvenile sta ge. oq, CM63079 (o), UMPC10210
(p), and UMPC7696 (q), all representing t he adult stage. I nterpretive dr awings
in bq are not to scale .
Extende d Data Fig. 9 | Se e next page for capti on.
Article
Extende d Data Fig. 9 | De terminati on of ontogen etic stage s for the
Palaeozoic stemlamprey specimens. This comparison reveals gradual
transiti ons of charact ers in each taxon , allowing the de signation of
ontogene tic stages and c orroboratin g their taxonom ic identity. Each
ontogene tic series of s tem lamprey shows an e arly expression of c haracter
states t hat, in modern lam preys, develop only in th e adult phase (dark blue
shades in cf), conf irming the ab sence of the am mocoete pha se in these ste m
taxa. a, Modern lampreys. Reference ontogenetic sequence and comparison of
ontogene tically variabl e characters t hat are also prese rved in the stem l amprey
specime ns, indicated i n blue shades. I mages are after a p revious public ation48
(adult, P. marinus) or courte sy of G. Kovalchuk (lat e larva and juvenile ,
Entosphenus tridentatus), except for the hat chling and early l arva (photogr aphs
by T.M., P. marinus). b, Size variations am ong the specim ens of P.riniensis.
Principa l component1 is us ed as a composit e size metric tha t correlates w ith
ontogenetic stages (xaxis), whereas the spec imens are similar in b ody length
from larval t o juvenile stage s (yaxis). For metho dological de tails, see the ‘A-3a.
Size variat ions’ secti on oftheSupplement ary Informati on. For original
measurem ents and princ ipal compone nt scores, see S upplementa ry Tables3
and 4, respec tively. cf, Ontogenet ic series of four Pa laeozoic stem lam preys:
P.riniensis (c); P.zangerli (d); M.pieckoensis (e); and H.montanensis (f). In
addition to t he character s that vary in mod ern lamprey ontoge ny (blue shades),
the table co mpares charac ters that var y ontogenetic ally in Priscomyzon (re d
shades) and in Hardistiella and Mayomyzon (gree n shades). Each row
represen ts a characte r; the lighter sha de shows the immatu re state (descri bed
in the left most column) and the d arker shade shows the m ature state
(describe d in the rightmo st column). Interm ediate shades i ndicate
polymor phism within th e stage (absence a nd presence mi xed) or interme diate
state (charac ter descri bes a continuu m). An empty box wit h a question mark
denotes m issing informat ion. Each sta ge represente d in the fossil record i s
linked to inter pretive drawing s of the specime ns (represent ative specime ns for
nonlarval s tages of Mayomyzon and Pipiscius). For detai led information a nd
discussi on of the charac ters, see the ‘A-3f. Defini tions of ontoge netic
charact ers’ sectio n oftheSupplementa ry Informatio n.
Extende d Data Fig. 10 | Phylo genetic tre e of early verte brate linea ges
resulti ng from our ana lysis, showi ng the relatio nships of st em lampreys
and other cyclostomes. The tree shows str ict consen sus of 360most
parsimoni ous trees bas ed on 52taxa and 167morpholo gical charac ters, revise d
from a previou s publication 10. Topology is similar to th e previously pre sented
consensus tree10, but differs i n: the deeply ne sted, stem-pe tromyzontiform
position o f Pipiscius; the stem-myx inoid aff inity of Gilpichthys; th e position of
Myxineidus as outg roup to all the rest of p etromyzonti forms; and the collap se
of crown node s into polyto mies for both myx inoids and petr omyzontiforms .
Filled circle s indicate crown n odes, and empt y circles show tot al nodes or the
most inclus ive nodes of enti rely extinct lin eages. Black , nonvertebra te
outgroup s; green, stem ver tebrates; ma genta, stem cycl ostomes; purp le,
anaspids (ste m cyclostomes); red, my xinoids (hag fishes); oran ge,
petromyz ontiforms (lamp reys); blue, gnathostom es (total group). For the
methodol ogy and analy tical detai ls, see the ‘Par t B’ secti on
oftheSupplementary Information.
... individual fossil specimens) are positioned closest to the extant adult semaphoront lamprey, while the Mesomyzon semataphonts are recovered close to the early decay stage extant adult semataphonts. The Priscomyzon fossil semataphont specimens previously interpreted as a range of ontogenetic stages [76] are all positioned near to the semataphonts representing decay stages of extant lampreys. The fossil specimens interpreted as later Priscomyzon ontogenetic stages ( juvenile and adults) are positioned closer to the extant adult lamprey semaphoront, and an early decay stage semataphont (stage 1) (figure 2c). ...
... Fossil Mayomyzon specimens comprise a morphological series that runs along the axis of decay-related variation observed in modern lamprey, with a few clear deviations ( figure 2b). A small number of Mayomyzon semataphonts, notably the specimen described as a late-larval stage with an advanced onset of decay (decay stage 4 (M.D4 [76]) (figure 2b)) occupy the same area as the early ontogenetic Priscomyzon specimens. There is a separation between the specimens described by Miyashita et al. [76] (largely interpreted as ontogenetic variation) and those from Gabbott et al. [31] (interpreted largely as taphonomic variation, driven by decay and preservational factors). ...
... A small number of Mayomyzon semataphonts, notably the specimen described as a late-larval stage with an advanced onset of decay (decay stage 4 (M.D4 [76]) (figure 2b)) occupy the same area as the early ontogenetic Priscomyzon specimens. There is a separation between the specimens described by Miyashita et al. [76] (largely interpreted as ontogenetic variation) and those from Gabbott et al. [31] (interpreted largely as taphonomic variation, driven by decay and preservational factors). While the Miyashita 'ontogenetic' specimens sit close to the extant adult semaphoront and Priscomyzon, the Gabbott specimens create a similar 'path' to the extant adult semataphonts, located between the third and sixth extant decay stages. ...
Article
Full-text available
Exceptionally preserved fossils of soft-bodied organisms provide unique evidence of evolutionary history, but they are often contentious; different approaches frequently produce radically different reconstructions of taxa and their affinities. Conflict arises due to difficulties in disentangling the three non-independent factors that underlie all morphological variation within and between fossils: ontogeny, taphonomy and phylogeny. Comparative data from extant organisms can be extremely powerful in this context, but is often difficult to apply given the multi-dimensionality of anatomical variation. Here, we present a multivariate ordination method using discrete morphological character data from modern taxa at different ontogenetic and taphonomic stages (semaphoront and 'semataphonts'). Analysing multiple axes of morphological variation simultaneously allows us to visualize character combinations that are likely to exist in fossil specimens at intersecting stages of growth and decay, and thus constrain interpretation of fossils. Application to early vertebrates finds variation in fossil specimens to be accounted for by all three axes: primarily decay in Mayomyzon, ontogeny in Priscomyzon and phylogeny in 'euphaneropoids' and Palaeospondylus. Our demonstration of empirical multi-factorial variation underscores the power of multivariate approaches to fossil interpretation, especially non-biomineralized taxa. As such, this conceptual approach provides a new method for resolving enigmatic taxa throughout the fossil record.
... In particular, we draw comparison to the morphologically and topologically comparable ventrolateral body-wall extensions in the trunk of cephalaspid osteostracans that are also continuous with the ventral lobe of the caudal fin from which they effectively bifurcate [19][20][21] . This structural and topological similarity, combined with the sister relationship between galeaspids and osteostracans plus jawed vertebrates 24,25 suggests that these structures are a shared characteristic retained in galeaspids and osteostracans, but lost in jawed vertebrates. The ventrolateral body-wall extensions do not extend to the branchial region in cephalaspids, nor do they overlap topologically with the paired pectoral fins. ...
... In any case, our insights into the postcranial anatomy of galeaspids allow us to constrain uncertainty over the intermediate sequence of evolutionary steps through which the monobasal paired fins of living jawed vertebrates evolved from the unpaired fins of the ancestral vertebrate 31,32 . Ancestral-state estimation using parsimony and likelihood optimization under alternative topologies of stem-gnathostome interrelations 24,25 suggests that paired ventrolateral body-wall extensions, as well as flexible paired fins, evolved early in the gnathostome stem lineage, before the divergence of osteostracans and jawed vertebrates (Fig. 4a). In contrast to the lateral fin-fold hypothesis, our results suggest that paired body-wall extensions were separated into pectoral and pelvic regions from the beginning, although this body plan was modified independently in numerous stem-gnathostome lineages, including galeaspids (Fig. 4b). ...
... We coded 55 taxa (18 extant and 37 extinct) for five discrete characters: (i) paired ventrolateral cephalothoracic dermal skeletal body-wall extensions, absent = 1, present = 2; (ii) paired ventrolateral fins, absent = 1, present = 2; (iii) extent of paired ventrolateral fins, absent = 1, continuous anteroposterior fins = 2, anterior fins only = 3, separate anterior and posterior fins = 4; (iv) pectoral appendicular endoskeleton and girdle, absent = 1, present = 2; (v) pelvic appendicular endoskeleton and girdle, absent = 1, present = 2. Character states were obtained from the literature (table 1 of Supplementary Data 3). Ancestral states were estimated using two phylogenetic hypotheses, either with anaspids as sister to all other ostracoderms + jawed vertebrates (as in ref. 24 ) or as sister to cyclostomes + conodonts (as in ref. 25 ). Interrelationships of heterostracans, thelodonts, osteostracans, placoderms and actinopterygians were based on refs. ...
Article
Full-text available
Galeaspids are extinct jawless relatives of living jawed vertebrates whose contribution to understanding the evolutionary assembly of the gnathostome bodyplan has been limited by absence of postcranial remains. Here, we describe Foxaspis novemura gen. et sp. nov., based on complete articulated remains from a newly discovered Konservat-Lagerstätte in the Early Devonian (Pragian, ∼410 Ma) of Guangxi, South China. F. novemura had a broad, circular dorso-ventrally compressed headshield, slender trunk and strongly asymmetrical hypochordal tail fin comprised of nine ray-like scale-covered digitations. This tail morphology contrasts with the symmetrical hypochordal tail fin of Tujiaaspis vividus, evidencing disparity in galeaspid post cranial anatomy. Analysis of swimming speed reveals galeaspids as moderately fast swimmers, capable of achieving greater cruising swimming speeds than their more derived jawless and jawed relatives. Our analyses reject the hypothesis of a driven trend towards increasingly active food acquisition which has been invoked to characterize early vertebrate evolution.
... Eriptychius fills an important gap, both temporal and phylogenetic, in our understanding of the evolution of the vertebrate head. Our inclusion of Eriptychius in a recent phylogenetic matrix for early vertebrates 45,46 recovers it within or in a polytomy with the vertebrate crown group and as a stem-group gnathostome in the Adams consensus of the parsimony analysis ( Fig. 3 and Extended Data Figs. 7-9) which is consistent with previous phylogenetic analysis 35 . ...
... 27,31 and supplementary appendix (p35) of ref. 45). The discovery of this preorbital neurocranium in Eriptychius and the movement of early vertebrate taxa around the vertebrate crown node in recent phylogenies 45,46 should prompt reconsideration of whether differences in orbital placement in Ordovician vertebrates instead reflect a more fundamental anatomical difference in cranial organization. ...
Article
Full-text available
The neurocranium is an integral part of the vertebrate head, itself a major evolutionary innovation1,2. However, its early history remains poorly understood, with great dissimilarity in form between the two living vertebrate groups: gnathostomes (jawed vertebrates) and cyclostomes (hagfishes and lampreys)2,3. The 100 Myr gap separating the Cambrian appearance of vertebrates4–6 from the earliest three-dimensionally preserved vertebrate neurocrania⁷ further obscures the origins of modern states. Here we use computed tomography to describe the cranial anatomy of an Ordovician stem-group gnathostome: Eriptychius americanus from the Harding Sandstone of Colorado, USA⁸. A fossilized head of Eriptychius preserves a symmetrical set of cartilages that we interpret as the preorbital neurocranium, enclosing the fronts of laterally placed orbits, terminally located mouth, olfactory bulbs and pineal organ. This suggests that, in the earliest gnathostomes, the neurocranium filled out the space between the dermal skeleton and brain, like in galeaspids, osteostracans and placoderms and unlike in cyclostomes². However, these cartilages are not fused into a single neurocranial unit, suggesting that this is a derived gnathostome trait. Eriptychius fills a major temporal and phylogenetic gap in our understanding of the evolution of the gnathostome head, revealing a neurocranium with an anatomy unlike that of any previously described vertebrate.
... L ampreys, together with hagfishes, are the only extant lineages of jawless fish 1,2 . Accumulating fossil evidence has demonstrated that lampreys in the Devonian period were already almost identical to the modern adult lampreys, with welldeveloped oral disc, annular cartilages, and circumoral teeth [3][4][5][6] , suggesting the evolutionary long-term stability of lampreys. ...
Article
Full-text available
Lampreys are blood-sucking vampires in marine environments. From a survival perspective, it is expected that the lamprey buccal gland exhibits a repository of pharmacologically active components to modulate the host’s homeostasis, inflammatory and immune responses. By analyzing the metabolic profiles of 14 different lamprey tissues, we show that two groups of metabolites in the buccal gland of lampreys, prostaglandins and the kynurenine pathway metabolites, can be injected into the host fish to assist lamprey blood feeding. Prostaglandins are well-known blood-sucking-associated metabolites that act as vasodilators and anticoagulants to maintain vascular homeostasis and are involved in inflammatory responses. The vasomotor reactivity test on catfish aortic ring showed that kynurenine can also relax the blood vessels of the host fish, thus improving the blood flow of the host fish at the bite site. Finally, a lamprey spatial metabolomics database ( https://www.lampreydb.com ) was constructed to assist studies using lampreys as animal model.
... Animal Husbandry. Sterlet sturgeon (Acipenser ruthenus Linnaeus, 1758) embryos and larvae were obtained from the Faculty of Fisheries and Protection (40) or (II) anaspids as stem cyclostomes (41,42). Based on this, the dermal armor was absent or secondarily lost in the cyclostome lineage (lampreys and hagfishes). ...
Article
Bone is an evolutionary novelty of vertebrates, likely to have first emerged as part of ancestral dermal armor that consisted of osteogenic and odontogenic components. Whether these early vertebrate structures arose from mesoderm or neural crest cells has been a matter of considerable debate. To examine the developmental origin of the bony part of the dermal armor, we have performed in vivo lineage tracing in the sterlet sturgeon, a representative of nonteleost ray-finned fish that has retained an extensive postcranial dermal skeleton. The results definitively show that sterlet trunk neural crest cells give rise to osteoblasts of the scutes. Transcriptional profiling further reveals neural crest gene signature in sterlet scutes as well as bichir scales. Finally, histological and microCT analyses of ray-finned fish dermal armor show that their scales and scutes are formed by bone, dentin, and hypermineralized covering tissues, in various combinations, that resemble those of the first armored vertebrates. Taken together, our results support a primitive skeletogenic role for the neural crest along the entire body axis, that was later progressively restricted to the cranial region during vertebrate evolution. Thus, the neural crest was a crucial evolutionary innovation driving the origin and diversification of dermal armor along the entire body axis.
... milii) basal to cartilaginous fishes, and reedfish (E. calabaricus) basal to bony fishes ( Fig. 2A) [22][23][24][25]. We also compiled representative orthologs of RAS proteins from key major groups of eukaryotes extending over 1000 MY of evolutionary distance, from single celled amoebozoans (D. discoideum) to mammals (H. ...
Article
Full-text available
KRAS, HRAS and NRAS proto-oncogenes belong to a family of 40 highly homologous genes, which in turn are a subset of a superfamily of >160 genes encoding small GTPases. RAS proteins consist of a globular G-domain (aa1-166) and a 22-23 aa unstructured hypervariable region (HVR) that mediates membrane targeting. The evolutionary origins of the RAS isoforms, their HVRs and alternative splicing of the KRAS locus has not been explored. We found that KRAS is basal to the RAS proto-oncogene family and its duplication generated HRAS in the common ancestor of vertebrates. In a second round of duplication HRAS generated NRAS and KRAS generated an additional RAS gene we have designated KRASBL, absent in mammals and birds. KRAS4A arose through a duplication and insertion of the 4th exon of NRAS into the 3rd intron of KRAS. We found evolutionary conservation of a short polybasic region (PBR1) in HRAS, NRAS and KRAS4A, a second polybasic region (PBR2) in KRAS4A, two neutralized basic residues (NB) and a serine in KRAS4B and KRASBL, and a modification of the CaaX motif in vertebrates with farnesyl rather than geranylgeranyl polyisoprene lipids, suggesting that a less hydrophobic membrane anchor is critical to RAS protein function. The persistence of four RAS isoforms through >400 million years of evolution argues strongly for differential function.
... The morphology of the hagfishes is regarded to be highly derived, including the degeneration of eyes because of adaptation to the deep sea (Dong & Allison, 2021;Gabbott et al., 2016), and many studies have instead used lampreys as a model for cyclostomes. However, recent studies have suggested that some lamprey characteristics that had been regarded as ancestral to the cyclostomes, such as the larval-type oral apparatus (Miyashita et al., 2021) and the transformation of the larval endostyle into the thyroid after metamorphosis (Takagi et al., 2022), are in fact the derived state from the acquisition of the ammocoete larval stage (still, careful discussion is necessary; see Mallatt, 2023). Lampreys and hagfish split around 470-390 million years ago (Kuraku & Kuratani, 2006; also see Miyashita et al., 2019), and their morphology has highly diverged. ...
Article
Full-text available
Hagfish (Myxinoidea) are a deep-sea taxon of cyclostomes, the extant jawless vertebrates. Many researchers have examined the anatomy and embryology of hagfish to shed light on the early evolution of vertebrates; however, the diversity within hagfish is often overlooked. Hagfish have three lineages, Myxininae, Eptatretinae, and Rubicundinae. Usually, textbook illustrations of hagfish anatomy reflect the morphology of the Myxininae lineage, especially Myxine glutinosa, with its single pair of external branchial pores. Here, we instead report the gross anatomy of an Eptatretinae, Eptatretus burgeri, which has six pairs of branchial pores, especially focusing on the coelomic organs. Dissections were performed on fixed and unfixed specimens to provide a guide for those doing organ- or tissue-specific molecular experiments. Our dissections revealed that the ventral aorta is Y-branched in E. burgeri, which differs from the unbranched morphology of Myxine. Otherwise, there were no differences in the morphology of the lingual apparatus or heart in the pharyngeal domain. The thyroid follicles were scattered around the ventral aorta, as has been reported for adult lampreys. The hepatobiliary system more closely resembled those of jawed vertebrates than those of adult lampreys, with the liver having two lobes and a bile duct connecting the gallbladder to each lobe. Overall, the visceral morphology of E. burgeri does not differ significantly from that of the known Myxine at the level of gross anatomy, although the branchial morphology is phylogenetically ancestral compared to Myxine.
Article
Full-text available
The vertebrate brain emerged more than ~500 million years ago in common evolutionary ancestors. To systematically trace its cellular and molecular origins, we established a spatially resolved cell type atlas of the entire brain of the sea lamprey—a jawless species whose phylogenetic position affords the reconstruction of ancestral vertebrate traits—based on extensive single-cell RNA-seq and in situ sequencing data. Comparisons of this atlas to neural data from the mouse and other jawed vertebrates unveiled various shared features that enabled the reconstruction of cell types, tissue structures and gene expression programs of the ancestral vertebrate brain. However, our analyses also revealed key tissues and cell types that arose later in evolution. For example, the ancestral brain was probably devoid of cerebellar cell types and oligodendrocytes (myelinating cells); our data suggest that the latter emerged from astrocyte-like evolutionary precursors in the jawed vertebrate lineage. Altogether, our work illuminates the cellular and molecular architecture of the ancestral vertebrate brain and provides a foundation for exploring its diversification during evolution.
Article
Tullimonstrum gregarium , also known as the Tully monster, is a well‐known phylogenetic enigma, fossils of which have been found only in the Mazon Creek Lagerstätte. The affinities of Tullimonstrum have been debated since its discovery in 1966, because its peculiar morphology with stalked eyes and a proboscis cannot easily be compared with any known animal morphotypes. Recently, the possibility that Tullimonstrum was a vertebrate has attracted much attention, and it has been postulated that Tullimonstrum might fill a gap in the fossil record of early vertebrates, providing important insights into vertebrate evolutionary history. With the hope of resolving this debate, we collected 3D surface data from 153 specimens of Tullimonstrum using a high‐resolution laser 3D scanner and conducted x‐ray micro‐computed tomographic (μCT) analysis of stylets in the proboscis. Our investigation of the resulting comprehensive 3D morphological dataset revealed that structures previously regarded as myomeres, tri‐lobed brain, tectal cartilages and fin rays are not comparable with those of vertebrates. These results raise further doubts about its vertebrate affinities, and suggest that Tullimonstrum may have been either a non‐vertebrate chordate or a protostome.
Preprint
Full-text available
KRAS, HRAS and NRAS oncogenes belong to a family of 40 highly homologous genes, which in turn are a subset of a superfamily of >160 genes encoding small GTPases. RAS oncoproteins consist of a globular G-domain (aa1-166) and a 22-23aa unstructured hypervariable region (HVR) that mediates membrane targeting. The evolutionarily origins of the RAS isoforms, their HVRs and alternative splicing of the KRAS locus has not been explored. We found that KRAS is basal to the oncogene family and its duplication generated HRAS in the common ancestor of vertebrates. In a second round of duplication HRAS generated NRAS and KRAS generated an additional RAS gene we have designated KRASBL, absent in mammals and birds. KRAS4A arose through a duplication and insertion of the 4th exon of NRAS into the 3rd intron of KRAS. We found evolutionarily conservation of a short polybasic region (PBR1) in HRAS, NRAS and KRAS4A, a second polybasic region (PBR2) in KRAS4A, two neutralized basic residues (NB) and a serine in KRAS4B and KRASBL, and a modification of the CaaX motif in vertebrates with farnesyl rather than geranylgeranyl polyisoprene lipids, suggesting that a less hydrophobic membrane anchor is critical to RAS oncoprotein function. The persistence of four RAS isoforms through >400 MY of evolution argues strongly for differential function.
Article
Full-text available
The Waterloo Farm lagerstätte in South Africa provides a uniquely well‐preserved record of a Latest Devonian estuarine ecosystem. Ecological evidence from it is reviewed, contextualised, and compared with that available from the analogous Swartvlei estuarine lake, with a particular emphasis on their piscean inhabitants. Although the taxonomic affinities of the estuarine species are temporally very different, the overall patterns of utilisation prove to be remarkably congruent, with similar trophic structures. Significantly, both systems show evidence of widespread use of estuaries as fish nurseries by both resident and marine migrant taxa. Holocene estuaries are almost exclusively utilised by actinopterygians which are overwhelmingly dominated by oviparous species. Complex strategies are utilised by estuarine resident species to avoid exposure of eggs to environmental stresses that characterize these systems. By contrast, many of the groups utilising Devonian estuaries were likely live bearers, potentially allowing them to avoid the challenges faced by oviparous taxa. This may have contributed to dominance of these systems by non‐actinoptergians prior to the End Devonian Mass Extinction. The association of early aquatic tetrapods at Waterloo Farm with a fish nursery environment is consistent with findings from North America, Belgium and Russia, and may be implied by the estuarine settings of a number of other Devonian tetrapods. Tetrapods apparently replace their sister group, the elpistostegids, in estuaries with both groups having been postulated to be adaptated to shallow water habitats where they could access small piscean prey. Correlation of tetrapods (and elpistostegids) with fish nursery areas in the Late Devonian lends strong support to this hypothesis, suggesting that adaptations permitting improved access to the abundant juvenile fish within the littoral zone of estuarine lakes and continental water bodies may have been pivotal in the evolution of tetrapods.
Article
Full-text available
Hagfish depart so much from other fishes anatomically that they were sometimes considered not fully vertebrate. They may represent: ( i ) an anatomically primitive outgroup of vertebrates (the morphology-based craniate hypothesis); or ( ii ) an anatomically degenerate vertebrate lineage sister to lampreys (the molecular-based cyclostome hypothesis). This systematic conundrum has become a prominent case of conflict between morphology- and molecular-based phylogenies. To date, the fossil record has offered few insights to this long-branch problem or the evolutionary history of hagfish in general, because unequivocal fossil members of the group are unknown. Here, we report an unequivocal fossil hagfish from the early Late Cretaceous of Lebanon. The soft tissue anatomy includes key attributes of living hagfish: cartilages of barbels, postcranial position of branchial apparatus, and chemical traces of slime glands. This indicates that the suite of characters unique to living hagfish appeared well before Cretaceous times. This new hagfish prompted a reevaluation of morphological characters for interrelationships among jawless vertebrates. By addressing nonindependence of characters, our phylogenetic analyses recovered hagfish and lampreys in a clade of cyclostomes (congruent with the cyclostome hypothesis) using only morphological data. This new phylogeny places the fossil taxon within the hagfish crown group, and resolved other putative fossil cyclostomes to the stem of either hagfish or lamprey crown groups. These results potentially resolve the morphological–molecular conflict at the base of the Vertebrata. Thus, assessment of character nonindependence may help reconcile morphological and molecular inferences for other major discords in animal phylogeny.
Article
Full-text available
Modern lampreys (Petromyzontiformes) are one of two lineages of surviving jawless fishes (agnathans), and are thus of critical importance to understanding the evolution of the vertebrates. Although their fossil record is meager, it appears they have remained morphologically conserved for at least 360 million years, but the origin of their multi-stage life history is unclear. Unlike hagfishes, the other extant group of jawless fishes, which exhibit direct development, all modern lampreys possess a complex life cycle which includes a long-lived freshwater larval (or ammocoete) period, followed by a true metamorphosis into a sexually-immature juvenile and then mature adult which differ dramatically in their morphology and ecology from the larva. Because of their basal position, it is critical to understand when the extant lamprey life history evolved, and if such a life history was present in the last common ancestor of agnathans and gnathostomes. Recent discoveries in paleontology, genomic analyses, and developmental biology are providing insight s into this problem. The current review synthesizes these findings and concludes that the ancestral lamprey life cycle followed a direct development. We suggest that the larval period was short and relatively limited if present at all, but that the juvenile included modern larval traits; over the course of evolution, differential select ion pressures throughout the lifetime produced distinct larval and juvenile/adult periods. Each period required the dramatically different morphologies seen in modern lampreys, ultimately requiring a true metamorphosis to accommodate the large changes in the body plan and to maximize the efficiency of each life period. As a result, modern lamprey life histories are a patchwork of ancestral and derived characters.
Article
Full-text available
Melanoma is a disease process which has been increasing in incidence over the past three decades and metastatic melanoma carries a poor prognosis. Through genetic studies of this disease, it has been determined that the BRAF V600 mutation plays a major role in the pathophysiology of the disease and this has led to the utilization of targeted therapy (BRAF and MEK inhibitors) in its treatment. Other BRAF mutations (non-V600 mutations) are rare in melanoma and targeted therapy is not indicated for patients with these mutations due to reduced response rates. An emerging option for metastatic melanoma with uncommon BRAF mutations is immunotherapy using checkpoint inhibitors such as PD-1 inhibitors or CTLA-4 inhibitors. Currently, it is unknown how patients with BRAF non-V600 mutations respond to immunotherapy. This report will examine the effect of immunotherapy on two distinct metastatic melanoma patients, each with uncommon BRAF mutations, occurring outside the V600 locus (E586K and G469E). These patients were noted to have a durable, complete response when treated with immunotherapy and continue to exhibit a response 9 and 15 months after discontinuing therapy. Further research and clinical trials are needed to study patients with uncommon BRAF mutations and the potential therapeutic benefit of immunotherapy.
Article
Full-text available
The success of vertebrates is linked to the evolution of a camera-style eye and sophisticated visual system. In the absence of useful data from fossils, scenarios for evolutionary assembly of the vertebrate eye have been based necessarily on evidence from development, molecular genetics and comparative anatomy in living vertebrates. Unfortunately, steps in the transition from a light-sensitive ‘eye spot’ in invertebrate chordates to an image-forming camera-style eye in jawed vertebrates are constrained only by hagfish and lampreys (cyclostomes), which are interpreted to reflect either an intermediate or degenerate condition. Here, we report—based on evidence of size, shape, preservation mode and localized occurrence—the presence of melanosomes (pigment-bearing organelles) in fossil cyclostome eyes. Time of flight secondary ion mass spectrometry analyses reveal secondary ions with a relative intensity characteristic of melanin as revealed through principal components analyses. Our data support the hypotheses that extant hagfish eyes are degenerate, not rudimentary, that cyclostomes are monophyletic, and that the ancestral vertebrate had a functional visual system. We also demonstrate integument pigmentation in fossil lampreys, opening up the exciting possibility of investigating colour patterning in Palaeozoic vertebrates. The examples we report add to the record of melanosome preservation in Carboniferous fossils and attest to surprising durability of melanosomes and biomolecular melanin.
Article
Full-text available
Reexamination of the chondrenchelyid Harpagofututor volsellorhinus from the Bear Gulch Limestone (Health Formation, Upper Chesterian, Mississippian), has revealed that the preserved fossil pigments are those originally associated with particular, well-vascularized abdominal organs (liver, spleen, and gonads) and major venous sinuses (orbital, gonadal, pelvic). The pattern of pigment localization reflects circulatory pathways of fossilized vessels. This determination was confirmed by comparison of fossil patterns to the visceral and circulatory anatomy of extant chondrichthyans. The arrangement of these pigments conveys strong, and otherwise unavailable, evidence for the internal reproductive features of these sexually mature, Paleozoic chondrichthyans. Under the appropriate preservational conditions the pigments also reveal asphyxia as the cause of death. Thus, the value of these pigments cannot be underestimated. Unfortunately, they are prone to spontaneous and progressive degradation that starts immediately upon excavation. Consequently, it is imperative to record data accurately and in a timely fashion. This report thus introduces the use of a color flatbed scanner as a particularly effective laboratory research tool for the collection and archiving of ephemeral fossil data.
Book
Distribution, variety and life cycles.- Perspectives and relationships.- Ecology and behaviour.- Respiration and feeding.- The heart and circulatory system.- The skeleton and the muscular system.- The nervous system.- Osmotic and ionic regulation.- The pituitary.- The peripheral endocrine tissues.- Reproduction and development.- Comparative biochemistry, immunology and cytogenetics.- Conclusions and evolutionary perspectives.