GEOL 431 Vertebrate Paleobiology

Spring Semester 2023

"God's noblest creation"

And now, a problem....

We have failed to address one sauropsid group that is dear to our institutional heart. We have saved Testudinata - turtles - for last because they have long represented the greatest challenge of any enigmatic vertebrate to phylogenetic systematics, despite their being very easy to recognize.

Pantestudines

(Permian - Quaternary) The total group containing Testudines and everything closer to them than to Lepidosauria or Archosauria. (Joyce et al., 2021.)

No vertebrate group is more radically transformed than turtles - a situation that has greatly hampered attempts to ascertain their phylogenetic relationships. We have even complained of lacking the Archaeopteryx of turtles, despite, apparently, having at least one under our noses for a century. Here we start with what's definitely known and work outward.

Major groups

(Joyce et al., 2021.)

• Pan-Testudines "The total clade of the crown clade Testudines" (In essence, Max ∇ (Testudines ~ Alligator, Lacerta))
• Testudinata "The clade for which a complete turtle shell, as inherited by Testudo graeca Linnaeus, 1758, is an apomorphy." (Effectively Min ∇ (Testudo, Proganochelys, Proterochersis))
• Testudines: Min ∇ (Chelonia, Trionyx, Chelonia, Testudo (cryptodires)). Includes:
  • Cryptodira: The mid-line head-retracting turtles - Min ∇ (Chelus, Pelomedusa, Podocnemis)
  • Pleurodira: The side-necked turtles - Min ∇ (Chelus, Kinosternon, Chelydra)

Major trends:

The turtle shell - Components: A composite structure in several senses:

• A dorsal carapace is integrated with a ventral plastron between the fore and hindlimbs. These sections form by distinct developmental processes and did not evolve simultaneously.

• Both carapace and plastron have:
  • a bony endochondral skeleton
  • a covering of broad keratinous scales.

  
  

  
  Keratinous (left) and bony (right) elements of the shell in snapping turtle, Chelydra serpentina
  Elements in both layers are identifiable and have names. Bony and keratinous elements do not match up one for one.

  
  

  
  Plastron elements from LafeberVet
• While the bones of the plastron are modified variants of those in other tetrapods:
  • Epiplastron = clavicle
  • Entoplastron = interclavicle
  • Remainder = gastralia

  
  

  
  Ventral view of carapace of Apolone spinifera (on its back)
  The elements of the carapace appear to be composites of modified endochondral ribs and vertebrae, and dermal bone of some sort. The elements:
  • Neurals incorporate dorsal vertebrae
  • Costals incorporate ribs
  • Marginals, suprapygal, and pygal have no obvious homologs
  • The nuchal's homologies are contentious (see below).

The turtle shell - Development:

How the components of the carapacial elements form and the precise identity of their homologs has been contentious. Major historic hypotheses include:

• The osteoderm hypothesis: maintains that the ribs and vertebrae fused developmentally with preexisting dermal scutes.
• The de novo hypothesis: holds that all of the shell elements are direct outgrowths of ribs and vertebrae.

Hirasawa et al., 2013, indicate a more subtle process.

in the costals:

• Ribs begin to form endochondrally, but are broadened by perichondral outgrowths.
• At the same time metaplastic (i.e. ectopic) bone forms in the adjacent dermis and fuses to the rib, yielding the platy costals.

Eventually, all suture to form a rigid shell. Thus, the development of the carapace is a little of both:

• The broadening of the ribs is absolutely a de novo process...
• But the metaplastic elements that fuse to them are, in essence, osteoderms.

In contrast to all other tetrapods:

• The ribs initially form in the somitic sclerotome but do not extend ventrally into the lateral plate mesoderm forming the body wall. Instead, they invade the dermis of the dorsal trunk.
• Rib formation is partially induced by the carapacial ridge that extends around the trunk. This feature functions like the apical ectodermal ridge of the limb buds, attracting the growing ribs and deflecting them from their normal developmental pathway. (Hirasawa et al., 2014)
• The elements of the plastron, in contrast seem to form from a secondary proliferation of neural crest (Clark et al., 2001.)

The lungs: Consequences of having a rigid shell that forms in this way:

• The girdles are enclosed inside the shell. (Not strictly inside the rib cage when one disregards shell components that aren't strictly homologous with ribs.)
• There is no room for intercostal muscles normally used to expand and contract the rib cage in breathing.
• Breathing is facilitated by the action of muscles on the:
  • posterior limiting menbrane posteriorly
  • and on the pectoral girdle anteriorly.

The girdles: As noted above:

• The turtle clavicle and interclavicle are incorporated into the plastron
• The cleithra are a puzzle:
  • Bever et al., 2015 and Lyson et al. 2016 maintain that they are fused to form the nuchal plate of the carapace, whereas...
  • Evidence for old-fashioned cleithra is reported for Kayentachelys (Joyce et al., 2006) and Eileanchelys (Anquetin et al., 2008)

this leaves the endochondral elements of the girdles.

• The pectoral girdle is triradiate with:
  • a long, slender scapular shaft,
  • a very long acromion process of the scapula,
  • and a broad anterior coracoid.
  Action of the serratus and pectoralis muscles rotates the girdle anteriorly, expanding the thoracic cavity.
• the pelvis connects to the carapace through two pairs of sacral ribs, and abuts the plaston ventrally. (Among pleurodires, it fuses to the shell top and bottom.)

The turtle skull: Characterized by widespread trends:

• Marginal and palatal teeth lost
• Jaws lined by a rhamphotheca (keratinous beak) as in birds and dicynodonts
• The adductor muscles of the jaw pass over the otic capsule, which acts like a pulley.
• The basipterygoid articulation is fused. (Compare with Petrolacosaurus.)
• The skull is anapsid, yet there is often (but not always) significant emargination of the temporal skull roof by embayments from the occipital region and the lower cheek margin. (An extreme case.)

Diversity of Pantestudines

Testudines:

(Late Jurassic - Quaternary) The crown group of living turtles encompasses two distinct clades:

• Cryptodira: the mid-line head-retracting turtles (familiar to North Americans)
• Pleurodira: the side-necked turtles (limited to the southern continents)

Note: Almost all living turtles retract their heads, but this capacity has evolved independently and happens in different ways. Earlier literature would include creatures from the Late Triassic and Early Jurassic, but recent work including Joyce, 2007 has booted these down onto the stem.

Synapomorphies of Testudines:

• Complete loss of palatal teeth
• Reduction of cervical ribs
• Reduction in tenth (posterior) dorsal rib
• Increase in number of pedal and manual phalanges

Pleurodira:

(side-neck - Late Jurassic - Quaternary) (above left)

Side-necked turtles retract their heads flexing their necks at the well developed articulations between cervical centra. Restricted to the southern (gondwanan) continents, they are the dominant turtles of Australia. Their synapomorphies include:

• Presence of a trochlear process of the pterygoid to guide the adductor muscles (right).
• Loss of the epipterygoid
• Presence of an anterior extension to the braincase wall
• Median contact of the posterior costals to form a suprapygal.
• Procoelous caudal vertebrae
• Pelvis solidly fused to shell

Pleurodiran superlatives:

• Stupendemys (Neogene of Venezuela) Fresh water aquatic. The largest turtle ever.
• Chelus fimbriata, the mata mata of South America. A cryptic turtle that has lost its beak and depends entirely on suction feeding.
• Chelodina longicollis, the Australian eastern long-necked turtle.

Cryptodira:

(Hidden neck - Late Jurassic - Quaternary) (above right)

The monophyletic group containing all turtles that pull their heads straight into their shells. Synapomorphies include:

• Presence of an anterior extension to the braincase wall (convergent with Pleurodira!)
• Pterygoid - basisphenoid contact conceals basipterygoid articulation.
• Retraction of neck along sagittal plane facilitated by accessory articulations of the neural arches

Noteworthy plesiomorphy: Otic capsule serves as trochlea for adductor muscles. Among living turtles, this distinguishes cryptodires, but it is also present among stem turtles.

Cryptodiran superlatives:

• Paracryptodira: (Late Jurassic - Paleogene) Near the cryptodire/pleurodire split, including the common Baenidae (right). Abundant fresh water turtles of the Cretaceous and Paleogene.
• Chelonioidea: (Cretaceous - Quaternary) Sea turtles, the only turtles successfully to invade the open oceans, including the largest living turtle (the leatherback) and Cretaceous superlatives like Protostega.
• Trionychia: (Cretaceous - Quaternary) Including pig-nosed turtles and softshelled turtles - fresh water turtles with reduced carapaces and limbs modified for rapid swimming.
• Testudinoidea: (Paleogene - Quaternary) Including tortoises and pond turtles.

Closest Relatives of Testudines:

The fossil record is full of creatures that are clearly turtles for all intents and purposes but fall outside the Testudines crown. Starting at the crown we move downward.

Meiolaniidae: (Cretaceous - Quaternary) Terrestrial turtles of the southern continents with:

• robust cranial horns and bosses
• long armored tails (independently derived)

The last species, Meiolania platyceps of Lord Howe Island became extinct only 2000 years ago.

Eileanchelys waldmani: (Middle Jurassic) Briefly noted - the first turtle with recognizable aquatic specializations.

Proganochelys: (Late Triassic) Turtles from farther down the tree are generally similar, with:

• Terrestrial adaptations (Illuminated by Joyce and Gauthier, 2004)
• Armored necks and tails

Of these, the earliest and best documented is Proganochelys (Late Triassic), well known from several specimens. It could not retract its legs, head, or tail into its shell. The neck and tail were studded with free-floating pieces of dermal armor. Nevertheless, it was turtle-like in overall appearance, with a full-blown shell and lacking marginal teeth.
It plesiomorphically displayed:

• Palatal teeth
• The absence of a trochlear system of the jaws
• A mobile basipterygoid articulation
• A clavicle (or is that "epiplastron?") with a dorsal process that reached the carapace.
• A relatively short acromion process.

Chinlechelys: (Late Traissic) Described by Joyce et al., 2008, the Late Triassic Chinlechelys of Arizona resembles Proganochelys except in that perichondral bone of its ribs and vertebrae are so loosely coossified with their dermal component, that their separate identity is clear. When first described, this seemed to resolve a long-standing enigma. In this early testudinatan, the skeletal elements of the shell seem to form by the coossification of the ribs with preexisting dermal plates, but it was not to be.

Testudinata: (Late Triassic - Quaternary) The last common ancestor of all creatures with complete turtle shells, and all of its descendants. Synapomorphies include:

• Ten dorsal vertebrate present
• Costal elements sutured together
• Nuchal, pygal, and costal elements present

Stem Testudinata: Now it gets interesting.

Odontochelys semitestacea:

Described by Li et al., 2008, the Late Triassic Odontochelys of China was at the time the most anatomically primitive turtle known. Indeed, it definitely lacks one synapomorphy - the toothless mouth. Odontochelys has typical simple teeth. Stranger yet, although it has a plastron, it lacks a proper carapace.

Distinctive features:

• Its depositional environment and limb shape indicates that it was aquatic.
• It's ribs are expanded into paddle-shapes, but
• there are no endochondral dermal scutes associated.

Odontochelys invokes a new big question: Did the plastron evolve before the carapace, or is Odontochelys secondarily derived in having lost its carapace?

• Some living highly aquatic turtles have reduced their carapaces significantly.
• However the plastron forms from a secondary proliferation of neural crest, a distinct developmental pathway from the carapace.

Synapomorphies with Testudinata:

• Paired gastralia transformed into plastron
• Addition of two cervical vertebrae
• incipient development of dorsal vertebrae into neural plates

Odontochelys was found in marine deposits of an embayment that was surrounded on three sides by land. Was it aquatic? That was the initial interpretation of Li et al., 2008 however it:

• is discordant with other animals in this part of the tree
• is not strongly supported anatomically. Morphometric comparative analyses by Joyce, 2015 and Dudgeon et al., 2021 maintain that the morphology of forelimbs in Odontochelys and other stem testudinatans supports terrestrially with, at most, occasional trips into the water.

Eorhynchochelys sinensis:

Li et al., 2018 Provide a glimpse of a slightly less derived version from the Early Late Triassic of China. Eorhynchochelys has the broadened ribs and flattened puboischiatic plate of turtles, but lacks any kind of plastron, offering a glimpse of the pantestudine condition just prior to that transformation.

But wait! Eorhynchochelys is derived in one respect: Although it retains maxillary teeth, the premaxillae and anterior dentaries seem to support a rhamphotheca!

Eunotosaurus africanus:

For a century, we have lamented not having the Archaeopteryx of turtles despite, evidently actually having it. Eunotosaurus africanus: (Middle Permian) Roughly a foot long, with a small head and roly-poly torso supported by expanded dorsal ribs. Known to science since 1892, this animal has alternately been considered close to and far from turtle ancestry.

Eunotosaurus shares with Odontochelys and Testudinata these synapomorphies:

• Reduced number of dorsal vertebrae
• Dorsal vertebrae elongate compared to cervicals and sacrals
• Dorsal ribs perichondrally expanded
• Judging from histology, intercostal muscles did not span the gaps between dorsal ribs.
• Trunk is wide
• Paired gastralia (however these are normal-looking)

Eunotosaurus is our first glimpse of the broadening of ribs. At this evolutionary stage, they offer little actual protection to the animal, but do:

• Reduce its stride length by making the trunk inflexible
• Reduce the ability to ventilate the lungs

What selective advantage encouraged their evolution? Lyson et al., 2016 argue that they are an adaptation to fossoriality, providing Eunotosaurus, with a solid base for the origin of powerful limb muscles.

Regardless of how phylogenetic analyses are structured, Eunotosaurus groups with Odontochelys and Testudinata.

Ontogeny and Phylogeny

One story that emerges from this phylogeny involves the position of the pectoral girdle with respect to the ribs. Ontogenetically, the scapula first forms in the normal position, but is able to slip beneath the carapace because the anterior ribs become angled posteriorly. Phylogenetically we see the same sequence:

• In Eunotosaurus the scapular blade is oriented dorsally anterior to the rib cage.
• In Odontochelys the dorsal process of the scapula is short and well anterior to the ribs because the distal ends of the ribs are squeezed together anteroposteriorly
• In Proganochelys the ribs (as costals) have resumed their normal orientation but the scapula is now ventral to them.

The Position of Pantestudines - May, 2015

Follow the bouncing turtles: When we try to say anything about the position of turtles in the tree of amniote evolution, things quickly get weird. Of all recent phylogenetic enigmas, turtles have been the most intractable.

Chronology of phylogenetic hypotheses:

• Gaffney and Meylan, 1988, deemed them to be related to captorhinids, basal eureptiles.
• Reisz and Laurin, 1991 indicated a possible close relationship between turtles and procolophonoids.
• Lee, 1993 proposed Pareiasaurs as a sister taxon. (In fact, according to later work by Lee, Pareiasaurs may be paraphylteic with respect to turtles, i.e. turtles may be a kind of pareiasaur.) Lee's argument assumed that the dermal component of the turtle shell derived from preexisting osteoderms.
• Rieppel and deBraga, 1996 performed a morphological cladistic analysis that placed turtles inside Lepidoauria, as members of Euryapsida, alongside sauropterygians. An interesting result for what it says about euryapsids! Traditionally turtle armor is seen as fundamentally different from that of armored placodonts but this result prompted open reconsideration.
• Lyson et al., 2010 re-ran the Rieppel and deBraga 1996 matrix adding Eunotosaurus and Proganochelys. The effect was that Eunotosaurus grouped as the sister taxon of Testudinata, inside Parareptilia.

Data type and results

Molecular analyses: For twenty years, relationships of Testudines to other extant groups have been addressed by molecular systematics. At various points they have been recovered as sister taxon to:

• Sauria
• Lepidosauria
• Archosauria
• Crocodylia
• Aves

By far the strongest pattern has been as sister to Archosauria, causing many authorities to accept the relationship. There are zero morphological synapomorphies uniting turtles and archosaurs.

The last decade: Other lines of evidence have been brought to bear on the issue of turtle relationships and the homologies of their skeletal elements.

• Werneburg and Sanchez-Villagra, 2009: Compiled an amniote phylogeny based on developmental timing features of organs and determined based on it that turtles are outside of Sauria. (Of course only living taxa could be considered in this way. Maybe outside of Diapsida.)

• Werneburg et. al, 2013 examined the development of the turtle neck, identifying embryonic intercentra and ribs that become coossified into cervical vertebrae. Among saurians, intercentra are greatly reduced, so their presence among turtles appears to argue for a non-saurian origin.

• The central and costal plates of the turtle shell are associated with vertebrae and ribs, respectively. Lyson et al., 2013 considered the homology of the enigmatic nuchal bone (right) , which has no vertebral component, and find it to be homologous to the cleithra of other amniotes. Recall that the cleithrum was the major bone of the pectoral girdle in fish-like sarcopterygians, reduced in amniotes, and lost in Sauria. The presence of the cleithrum in turtles, again, argues against their being members of Sauria.

2015 - Summer of Revelation:

When we taught GEOL431 in 2015, the accumulating morphological and developmental data seem to support the parareptilian hypothesis of turtle origins. Then came an interesting summer:

Pappochelys rosinae (Middle Triassic of Germany) was described by Schoch and Sues, 2015. This small (~30 cm overall length) animal is intermediate between Eunotosaurus and Odontochelys both in time and morphology:

• Its ribs are expanded but there are no dermal scutes associated.
• Its gastralia are very robust but not developed into a plastron.

They are clearly closely related.

But its remarkable feature is its diapsid skull, that with its ventrally open infratemporal fenestra resembles that of a derived stem saurian. It appears that turtles are, in fact, some kind of diapsid. But what about Eunotosaurus? It has an anapsid skull, right?

Not actually. The skull roof of Eunotosaurus is covered with dermal bone, however it has an odd feature: the supratemporal projects far anteriorly. By digitally removing it form high-resolution CT scans, Bever et al., 2015 revealed that it covered a proper supratemporal fenestra. Moreover, they showed ontogenetic variation in the development of the anterior extension of the supratemporal bone. Juveniles, it developed, had proper supratemporal fenestrae.

Why would Eunotosaurus do this?

• The strengthening of the skull is common in fossorial animals. Seeing this in Eunotosaurus is consilient with Lyson et al.'s, 2016 argument for fossoriality.

And what does this tell us about the importance of fenestration patterns as "key characters" in taxonomy? :-(

Caveat:

But note, in some ways Eunotosaurus really isn't what we expect of saurian or nearly saurian turtle precursor:

• Its rod-shaped squamosal is reminiscent of that of a procolophonoid parareptile and is dissimilar to the four-pronged squamosals of saurians
• It lacks a suborbital fenestra in the palate
• Its ribs imbricate with one another (overlap) in a manner unlike other stem testudinatans

Indeed, Lichtig and Lucas, 2021, in their redescription of Chinlechelys, reaffirm that turtles belong in Parareptilia, and denounce Eunotosaurus and Pappochelys as red-herrings that are not close turtle relatives. We still await consensus.

So what are they?

Do we just shrug our shoulders and admit that turtles belong in Archosauria? Not yet! Two issues to consider:

• Pappochelys is informally reported (Sues, 2015 pers. comm.) to have:
  • a facet for the cleithrum on its scapula. Moreover, evidence for cleithra is reported for Kayentachelys (Joyce et al., 2006) and Eileanchelys (Anquetin et al., 2008) as well.
  • no descending flanges of the parietal

  That indicates that although they may be diapsids, they are outside of Sauria.

• Finally, no morphological data can really refute the molecular data that suggest that turtles are archosaurs. Indeed, it is difficult to compare these data directly.

We hope for salvation through future revelations. Until then, Schoch and Sues, 2019 provide an excellent review.

Amen.

Eunotosaurus skull reconstruction from Bever et al., 2015.