Loganellia scotica: even more like a tiny whale shark

For over a century
taxon omission overlooked this homology.

Figure 1.  Manta compared to Thelodus (Loganellia) and Rhincodon. All three have a terminal mouth essentially straight across, between the lateral eyes, distinct from most fish.

Figure 1.  Manta compared to Thelodus (Loganellia) and Rhincodon. All three have a terminal mouth essentially straight across, between the lateral eyes, distinct from most fish.

Earlier I added
the Silurian thelodont jawless fish, Thelodus (Fig. 2), as the outgroup for the large reptile tree (LRT, 1597+ taxa; subset Fig. 6). The whale shark (Rhincodon) and the manta ray (Manta) (Fig. 1) nested as the first outgroup taxa. In my mind there is a little confusion over which specimens belong to Thelodus and which belong two Loganellia or otherwise.

Figure 2. Thelodont with whale shark shape including dorsal fin. Image from OldRedSandstone.com. This appears to be Loganellia, not Thelodus (Fig. 7).

Figure 2. Thelodont with whale shark shape including dorsal fin. Image from OldRedSandstone.com. This appears to be Loganellia, not Thelodus (Fig. 7).

After seeing an online photo
of another Silurian thelodont, Loganellia scotica (Traquair 1898, Early Silurian, 430 mya; Fig. 3), I was able to draw out more details than the traditional line drawing offered using DGS methodology. Many similar Silurian thelodonts are known (Fig. 2). Some, evidently, have changed names over the last 120 years, probably because details are difficult to glean. I have not been able to ascertain any museum numbers for these taxa.

Figure 1. Loganellia scotica in situ and after DGS coloring. Several details are recovered here that identify this as a little sister to Rhincodon, the whale shark. Photo from Birkknowesaffairs.com. Specimen from museum?

Figure 3. Loganellia scotica in situ and after DGS coloring. Several details are recovered here that identify this as a little sister to Rhincodon, the whale shark. Photo from Birkknowesaffairs.com. Specimen from museum?

After DGS, Loganellia appears closer to Rhincodon
than the traditional ink diagram by Traquair (Figs. 1, 3) indicates. Pelvic fins are present. One or two dorsal fins are present. An anal fin and cutwaters are present. A mandible is present. That comes as a surprise for a supposedly jawless taxon.

Rücklin et al. 2019 studied Loganellia
and reported, “The available evidence suggests that internal odontodes evolved through the expansion of odontogenic competence from external to internal epithelia.” Neither Rhincodon nor Manta are found in the text. Both of these taxa have the sort of scale-like tooth carpet the authors were able to see using synchrotron radiation X-ray tomographic microscopy (SRXTM). They report, “We reveal that the internal scales are organized into fused patches and rows, a key distinction from the discrete dermal scales. 

Figure 1. Whale shark (Rhincodon) tooth pads, not that much different from catfish tooth pads (Fig. 2).

Figure 4. Whale shark (Rhincodon) tooth pads, similar to those described for Loganiella.

Rücklin et al. continue:
“Internal scales, where present, are always located near to external orifices; the sequential development of pharyngeal scales in Loganellia is peculiar among thelodonts and other stem gnathostomes. It represents a convergence on, rather than the establishment of, the developmental pattern underpinning tooth replacement in jawed vertebrates.”

Perhaps Rücklin et al. were influenced by
the traditional hypothesis that placoderms were the first jawed vertebrates, derived from osteostracan jawless vertebrates (Fig. 3). The LRT (subset Fig. 6) does not support the traditional hypothesis.

Figure 4. From Zhu et al. 2016, overlay added based on LRT topology.

Figure 5. From Zhu et al. 2016, overlay added based on LRT topology.

Figure 9. Subset of the LRT focusing on fish. The two nominal Gemuendia specimens are highlighted.

Figure 6. Subset of the LRT focusing on fish. The two nominal Gemuendia specimens are highlighted.

A final note:
Long 1995 identified this specimen (Fig. 7) as Thelodus. Distinct from Loganellia (Fig. 3), this specimen assigned to Thelodus appears to have an underslung jaw and long, raking teeth, perhaps to sift through seafloor sand, long pectoral fins and other distinct traits. Some thelodonts do not have fins and jaws.

Figure 7. Specimen identified as Thelodus by Long 1995. It has an underslung mandible and long rostral teeth along with long pectoral fins, distinct from Loganellia.

Figure 7. Specimen identified as Thelodus by Long 1995. It has an underslung mandible and long rostral teeth along with long pectoral fins, distinct from Loganellia.

References
Long JA 1995.The Rise of Fishes. 500 million years of evolution. The Johns Hopkins University Press, Baltimore and London.
Rücklin M, Giles S, Janvier P and Donoghue PCJ 2011.
Teeth before jaws? Comparative analysis of the structure and development of the external and internal scales in the extinct jawless vertebrate Loganellia scotica. Evolution & Development 13(6):523–532.
Traquair RH 1898. Report on fossils fishes. Summary of Progress of the Geological Survey of the United Kingdom for 1897: 72-76.
Žigaite· Ž & Goujet D 2012. New observations on the squamation patterns of articulated specimens of Loganellia scotica (Traquair, 1898) (Vertebrata: Thelodonti) from the Lower Silurian of Scotland. Geodiversitas 34 (2): 253-270.

wiki/Loganellia
oldredsandstone.com

Putting two new faces on Gemuendina

Thanks to DGS
there’s more to know about this strange bottom-dwelling fish taxon, die Panzerfisch.

The classic view
of Gemuendina stuertzi (Traquair 1903; Figs. 1-4); Early Devonian; 15cm to 1m in length) nests this taxon among the placoderms. A drawing from Gross 1963 (Fig. 1) makes it look like a scaly ray or a stargazer (a type of goosefish that buries itself in sand waiting for prey to swim by). Regrettably, very few of the bones / scales in the Gross diagram look like those of any other taxon in the large reptile tree (LRT, 1597 taxa). So, that means we have some work to do to figure this thing out.

Figure 1. Based on the scale bar, this a surprisingly small specimen, full scale on a typical computer monitor.

Figure 1. Based on the scale bar, this a surprisingly small specimen, full scale on a typical computer monitor.

Wikipedia reports,
“Unlike most other placoderm orders Gemuendina and its four other known relatives had armor made up of a mosaic of unfused bony plates and scales.”

That sounds like Stensioella (Fig. 5), another purported placoderm that instead nests closer to lungfish in the LRT.

Wikipedia reports,
“Also unlike other placoderms, it did not have the characteristic tooth plates of placoderms. Instead, it had star-shaped tubercle scales that allowed it to seize, then swallow fish and other animals that swam too close with its mouth.”

Several specimens are referred to this genus. Most are less complete than the one pictured here (Fig. 2). I was unable to figure out which one was the holotype.

Figure 2. Gemuendina in situ with DGS colors applied to scattered skull bones and gill covers. Light red curved bones represent the pectoral girdle, not ray-like pectoral fins. This may have been less ray-like than traditionally portrayed.

Figure 2. Gemuendina in situ with DGS colors applied to scattered skull bones and gill covers. Light red curved bones represent the pectoral girdle, not ray-like pectoral fins. This may have been less ray-like than traditionally portrayed.

Coloring the details in Gemuendina
(Fig. 2), reveals a different sort of fish when reconstructed (Fig. 3) than traditionally portrayed (Fig. 1). Here the traditional skull bones appear. Many are displaced from their in vivo positions. Here (Fig. 3) most of the skull bones are back in place.

Figure 3. Gemuendian reconstructed. The gray shapes are gill cover plates. The rest of the bones nest Gemuendina closer to the living bichir, Polypterus. This may not be the holotype specimen, but an improperly referred specimen. Not sure yet.

Figure 3. Gemuendian reconstructed. The gray shapes are gill cover plates. The rest of the bones nest Gemuendina closer to the living bichir, Polypterus. This may not be the holotype specimen, but an improperly referred specimen. Not sure yet.

In the LRT this rather complete (missing the tail tip) specimen
of Gemuendina (Fig. 3) nests it in the LRT with the extant bichir, Polypterus (Fig. 4) far from the placoderms. Rebuilding the loose skull bones using DGS made this possible.

Figure 1. The Nile bichir (Polypterus), skull, skeleton and bones colorized for ease of comparison. Compare to the placoderm, Entelognathus, (Fig. 2) and the stem tetrapod Tinirau (Fig. 3).

Figure 4. The Nile bichir (Polypterus), skull, skeleton and bones colorized for ease of comparison.

And also
with the basically similar basal lobefin, Stensioella (Fig. 5).

Figure 5. Stensioella in situ and colorized. This taxon is also not a placoderm in the LRT.

Figure 5. Stensioella in situ and colorized. This taxon is also not a placoderm in the LRT.

I’ve avoided Gemuendina until now
because the traditional diagram (Fig. 1) seemed so different. Now I know why. There’s too much imagination in that freehand drawing. Tracing is so much better, even if less esthetic.

But wait! There’s more.
Other, less complete specimens (Figs. 6, 7) are also referred to Gemuendina.

Figure 6. Specimen KGM-1983_308 has a dorsal naris with closely appressed openings.

Figure 6. Specimen KGM-1983_308 has a dorsal naris with closely appressed openings. The mandible is in blue. Premaxilla in yellow. Nasals in pink.

Just because they’re flat and wide,
and more or less articulated, these specimens also need to be understood bone-by-bone  before referring them to the taxon wastebasket that is Gemuendina. If anyone knows which specimen is the holotype, that will clear some things up. The other specimens will then need new names.

Figure 7. the KGM 1983 306 specimen referred to Gemuendina. This one is closer to the extant channel catfish, Ictalurus.

Figure 7. the KGM 1983 306 specimen referred to Gemuendina. This one is closer to the extant channel catfish, Ictalurus.

The KGM 1983 306 specimen
nests in the LRT with the extant channel catfish, Ictalurus. Catfish have the primitive pavement of teeth seen in the KGM 1983 306 specimen. Evidently this was overlooked.

Figure 8. The extant channel catfish, Ictalurus. Note the lack of cheekbones, maxilla and lacrimal.

Figure 8. The extant channel catfish, Ictalurus. Note the lack of cheekbones, maxilla and lacrimal.

Both catfish sister taxa lack cheekbones
(jugal + squamosal), maxilla and lacrimal. Catfish, as we learned earlier, are sisters to placoderms in the LRT. That’s why Gemuenida is often referred to as a different sort of placoderm. Now we have an Early Devonian catfish, KGM 1983 306, the first ever identified from that time period. (Let me know if this was discovered earlier and I will make the citation known.)

Figure 9. Subset of the LRT focusing on fish. The two nominal Gemuendia specimens are highlighted.

Figure 9. Subset of the LRT focusing on fish. The two nominal Gemuendia specimens are highlighted.


References
Gross W 1963. Gemuendina stuertzi Traquair. Notizblatt des Hessischen Landesanstalt fu¨r Bodenforschung 91, 36–73.
Traquair RH 1903. The Lower Devonian Fishes of Gemunden. Transactions of The Royal Society of Edinburgh 40:723-739.

wiki/Gemuendina

Fish evolution: Semionotus to Lepisosteus

Convergent with the stickleback/seahorse clade
these taxa shift the jaw joint anterior to the orbit, evolve a longer rostrum and have an armored body.

Figure 1. The extant gar (Lepisosteus) and its Late Triassic ancestor (Semionotus).

Figure 1. The extant gar (Lepisosteus) and its Late Triassic ancestor (Semionotus).

Putting sister taxa together graphically
shows how evolution turned a shorts-snouted Late Triassic armored fish into a long-nosed extant gar, reflecting their nesting in the large reptile tree (LRT, 1595+ taxa, Fig. 3).

Figure 2. Skulls of the extant gar (Lepisosteus) and its Late Triassic ancestor (Semionotus).

Figure 2. Skulls of the extant gar (Lepisosteus) and its Late Triassic ancestor (Semionotus).

Mea culpa
Earlier I nested the gar, Lepisosteus) with the stickleback, Gasterosteus. In order to do that I gave different identities to the skull bones than shown here (Fig. 2). At the time there were no other taxa with the jaw joint anterior to the orbit. Swordfish nested nearby and their juveniles have long, toothy jaws. So it all made sense. Now I see that Semionotus has the same arrangement. Both are traditional holosteans. Other traditional holosteans, Pseudoscaphirhynchus, the sturgeon, and Amia, the bowfin continue to nest outside the rest of the Holostei (Fig. 2). Adding taxa (not characters) solves all such problems.

Semionotus bergeri (Agassiz 1843; Late Triassic) also has thick scales and small fins. Here the maxilla is even smaller.

Lepidosteus osseus (also Lepisosteus Lacepéde 1803) is the extant longnose gar. The rostrum is elongated. The mandible is anterior to the orbit, as in the stickleback, Gasterosteus. Ganoid scales cover the body.

Figure 3. Subset of the LRT focusing on fish. This reflects a change from previous cladograms.

Figure 3. Subset of the LRT focusing on fish. This reflects a change from previous cladograms.

This is a new cladogram of fossil fish
representing a new hypothesis of interrelationships.


References
Agassiz L 1843. Poissons Fossil 833.
de Lacepéde BG 1803. Histoire naturelle des poissons. Tome Cinquieme. 5(1-21):1-803 + index.

wiki/Lepidosteus
wiki/Semionotus

SVP abstracts – JAWS! (+ tooth genesis)

Chen et al. 2019 bring us a new look
at the first teeth and jaws to evolve in the clade Gnathostomata.

A word of caution:
Watch out for, and be able to differentiate between ‘the origin of jaws and teeth’ vs ‘the reduction of jaws and teeth’ here. The two can be confused with one another if you don’t have a good outgroup taxon.

Therefore it is helpful to start with a jawless outgroup
for polarity. In the large reptile tree, (LRT, 1593 taxa; subset Fig. 6) the thelodont, Thelodus (Fig. 1), is that outgroup taxon. In other words, Thelodus is the last common ancestor of all vertebrates with jaws in the LRT. That means some undiscovered taxon or Thelodus itself, documents the genesis of jaws. Since all Thelodus specimens are crushed flat body fossils with no internal details shown or known, we’ll have to wait to find out what is inside some to-be-discovered 3D Thelodus. Perhaps the transition occurs within that genus. Meanwhile…

Figure 4. Manta compared to Thelodus and Rhincodon. All three have a terminal mouth essentially straight across, between the lateral eyes.

Figure 1. Manta compared to Thelodus and Rhincodon. All three have a terminal mouth essentially transverse between the lateral eyes set far forward on the skull. Based on Thelodus, the outgroup, this is the primitive condition.

From the abstract:
“Osteichthyan dentitions are characterized by cyclic tooth replacement and linear tooth rows.”

Not all of them. Whale sharks (Fig. 1; Rhincodon), manta rays (Fig. 1; Manta ) have tooth carpets and no marginal teeth on the jaws (Fig. 1).

Figure 1. Whale shark (Rhincodon) tooth pads, not that much different from catfish tooth pads (Fig. 2).

Figure 2. Whale shark (Rhincodon) tooth pads, not that much different from catfish tooth pads.

Chen et al. continue:
“The acquisition of these characters can be explained by an intimate relationship between the growth of jawbone and the initiation of teeth, supported by substantial evidence from synchrotron microtomography that reveals the 3D pattern of successor teeth, vascular canals, growth arrested and resorption surfaces in various Silurian-Devonian gnathostomes.

Jawbones (premaxilla, maxilla and dentary) are actually the last bones to get teeth. So, according to the LRT, the process of getting marginal teeth was a little more complicated than Chen et al. proposed. Perhaps it happened twice by convergence. Catfish (Fig. 3) show an early version of this in bony fish. Sharks evolved marginal teeth by convergence.

Figure 4. Catfish teeth from Usman et al. 2013, colors added.

Figure 3. Catfish teeth from Usman et al. 2013, colors added. The labeled maxillary teeth are actually premaxillary teeth. Compare these carpets of short teeth with those in figure 2.

Chen et al. continue:
The growing bone provides space for new teeth to attach and the succession of larger teeth maintains the number of teeth as the animal grows.”

OK. Gong with the flow, let’s just focus on jawbones now.

“In non shedding dentitions, whether the spiral addition of acanthodian tooth whorls, the anterior addition of ischnacanthid dentigerous jawbones, or the radial addition of arthrodire gnathal plates, the sequential addition of teeth is synchronized with the appositional growth of bone.”

Figure 2. Onychodus and Ischnacanthus share enough traits to make them sisters, apart from Brachyacanthus + Pteronisculus.

Figure 4. Onychodus and Ischnacanthus share enough traits to make them sisters, apart from Brachyacanthus + Pteronisculus. Circles show the acanthodian tooth whorls mentioned above.

Chen et al. continue:
The most primitive teeth of the most basal stem gnathostome Radotina and Kosoraspis already display the lingual addition of tooth rows shared by the gnathostome crown group.”

The LRT recovers more primitive taxa with teeth on the jaws, like Squatina, the extant angel shark (Fig. 5). It is basically a large Thelodus. You know it has primitive jaws because surrounding the jaws on either side are gill bars. These slowly disappear in all more derived taxa. Radotina and Kosoraspis are both Early Devonian placoderms, a derived gnathostome clade (Fig. 6). Neither preserve lateral gill bars.

On a side note, and regarding placoderms, Vaskaninova and Ahlberg 2017 wrote, “A key development in the understanding of this stem group has been the recognition that the placoderms (armoured jawed fishes of Silurian to Devonian age), which until recently were regarded as a clade branching off the gnathostome stem group, probably form a paraphyletic segment of that stem group. Some groups of placoderms appear to be very primitive and close to jawless vertebrates whereas others possess what were previously regarded as osteichthyan autapomorphies (notably a maxilla, premaxilla and dentary) and are probably close to the gnathostome crown-group node.”

The LRT does not support these placoderm hypotheses. All tested placoderms nest together in the LRT at the base of the lobefin-rayfin split. Some bottom dwelling placoderms revert to a near jawless condition. So do certain catfish, like the kind that cling to aquarium glass.

Figure 5. Squatina skull. Note the gill bars framing the mouth. These are modified in Aetobatus into a digging snout.

Figure 5. Squatina skull. Note the gill bars framing the mouth and the marginal teeth, the most primitive examples in the shark lineage. Heterodontus has the most primitive teeth in the bony fish line, despite the lack of bones in Heterodontus.

Chen et al. continue:
When in situ resorption evolved in osteichthyans, the first-generation teeth of the stem osteichthyan Lophosteus are shed semi-basally forming deeply overlapping tooth rows. As the bone growth slows down at later developmental stages, the succeeding teeth overlap the preceding ones entirely, causing the preceding teeth to be shed basally and replaced in situ. 

“Therefore, tooth replacement may have emerged via a tooth initiation rate higher than the bone growth rate. When a lingual shelf is formed on the marginal jawbones of crown osteichthyans, the lingual growth of bone is restricted, and new tooth rows cannot be added lingually to the previous rows, only apically.”

In the LRT, other than sharks, the most primitive instance of marginal teeth occurs in chimaeras like Falcatus (Fig. 7).

Figure 1. The skull of Falcatus with DGS tracing above.

Figure 7. The skull of Falcatus with DGS tracing above. Note the primitive marginal teeth.

Chen et al. continue:
“The replacement of the marginal linear tooth row of the basal actinopterygian Moythomasia is actually a vertical piling of alternate tooth rows by semi-basal resorption. Thus a single linear tooth row may have transformed from a lingual-labial compressed version of transverse tooth files.”

Moythomasia was added to the LRT, but it nests off a rather derived node, far from the most primitive taxa in the LRT with jaws. Like Falcatus (Fig. 7), Moythomasia has deep jaws that extend to the back of its skull. By contrast the whale shark and other basal taxa have much shallower, more transverse jaws, more like those of Thelodus (Fig. 1).


References
Chen et al.  2019. Development relationships between teeth and jawbones in stem gnathostomes and stem osteichthyans revealed by 3D histology: insights into the evolution of tooth replacement and tooth organization.
Vaskaninova V and Ahlberg PE 2017. Unique diversity of acanthothoracid placoderms (basal jawed vertebrates) in the Early Devonian of the Prague Basin, Czech Republic: A new look at Radotina and Holopetalichthys. PLoS ONE 12(4):eo174794.

Gregorius rexi: not a ratfish in the LRT

Revised November 07, 10 and 17 2019
with a revision to the LRT that moves Gregorius closer to Hybodus, basal to Placodermi.

Talk about a transitional taxon…
Gregorius rexi (Lund and Grogan 2004; 11cm long; Early Carboniferous; CM 35490) is a small fish from the famous Bear Gulch Formation in Montana. Traditionally it is considered a type of ratfish.

By contrast,
in the large reptile tree (LRT, 1593 taxa; Fig. 3), Gregorius is a late surviving member of an Early Devonian genesis representing the most primitive ray-fin fish splitting from Hybodus. Gregorius is the last common ancestor of all bony fish and placoderms. Ratfish are a bit more primitive.

Figure 1. Gregorius rexi enlarged and to to scale with its cousin in the LRT, Robustichthys. Gregorius still has a dorsal spine and an odd soft of diphycercal tail.

Figure 1. Gregorius rexi enlarged and to to scale with its cousin in the LRT, Robustichthys. Gregorius still has a dorsal spine and an odd soft of diphycercal tail.

Gregorius is not far from catfish,
still tucked inside the placoderms. Thunnus, the tuna, is the most primitive extant ray-fin fish among taxa derived from a sister to Gregorius.

FIgure x. Gregorius rexi with articulated bones in vivo.

Figure 2. Gregorius rexi with articulated bones in vivo. This taxon is basal to the clade Placodermi. 

In the meantime,
I’ve been learning more about ray fin fish. Some taxa have moved around as mistakes are discovered and corrections are made. The four-eyed fish now nests with the mudskipper, as an example. The general topology of the tree has otherwise stayed much the same as the Bootstrap scores get better. When that portion is complete, we’ll review the changes.

Figure 9. Subset of the LRT focusing on fish. The two nominal Gemuendia specimens are highlighted.

Figure 9. Subset of the LRT focusing on fish. The two nominal Gemuendia specimens are highlighted.


References
Lund R and Grogan E 2004. Five new euchondrocephalan Chondrichthyes from the Bear Gulch Limestone (Serpukhovian, Namurian E2b) of Montana, USA. Recent Advances in the Origin and Early Radiation of Vertebrates 505-531.

https://people.sju.edu/~egrogan/BearGulch/pages_fish_species/Gregorius_rexi.html

Tooth whorls: Helicoprion, Ischnagnathus and Onycodontus

After decades of wondering and guessing,
Tapanila et al. 2013 provided µCT scans of the enigmatic basal vertebrate (fish) Helicoprion vessonowi (Figs. 1-3; Karpinsky 1899; Permian, 290-270 mya; possibly 12m in length). Helicoprion is represented by several fossils whorls of teeth of decreasing size over several spiraling revolutions representing growth of this whorl without tooth loss. Often enough such tooth whorls are found in phosphate mines.

https://www.wired.com/2011/03/unraveling-the-nature-of-the-whorl-toothed-shark/

FIgure 1. Helicoprion fossil. Evidently this was an adult based on the number of smaller teeth present here.

According to Wikipedia,Helicoprion is a genus of extinct, shark-like eugeneodontid holocephalid fish.” Ratfish are considered the closest extant relatives. Chimaera is the closest tested taxon in the large reptile tree (LRT, 1592 taxa). No other eugeneotontids have been tested yet.

According to Brian Switek writing for NatGeo.com,
“A very special fossil – IMNH 37899 – preserved both the upper and lower jaws in a closed position, finally solving the mystery of what the ratfish’s head actually looked like. But determining the exact placement of that vexing spiral was just an initial step.”

As you can see
(Figs. 2, 3), traditional reconstructions over the past 6 years have featured a long-snouted shark-like form, based on the data in Tapanila et al. The evidence indicates that the only portion of the skull (sans mandible) recovered in the µCT scans is a narrow set of palate cartilage (pterygoids + palatines), a narrow set of mandibles and a cartilage plate covering the center of the whorl. Despite everyone from Tapanila et al. to Switek calling this a ratfish, paleoartists keep providing a shark-like image (Fig. 2) lacking pelvic fins. Sharks typically have a wider skull. Ratfish skulls (Figs. 2, 3) are typically taller and narrower, providing a closer match to the recovered palate and mandible shapes. Tapanila et al. regarded Helicoprion as a member of the Holocephalia, the clade of ratfish (see below). Finally note the jaws cannot completely close because the tooth whorl gets in the way.

Figure 1. Helicoprion µCT scans, model made from scans, in vivo image with shark-like proportions, and matched against chimaera (Hydrolagus) jaws.

Figure 2. Helicoprion µCT scans, model made from scans, in vivo image with shark-like proportions, and matched against chimaera (Hydrolagus) jaws. See figure 2 for closeup.

From the Tapanila et al. abstract, 
“New CT scans of the spiral-tooth fossil, Helicoprion, resolve a longstanding mystery concerning the form and phylogeny of this ancient cartilaginous fish. We present the first three-dimensional images that show the tooth whorl occupying the entire mandibular arch, and which is supported along the midline of the lower jaw. Several characters of the upper jaw show that it articulated with the neurocranium in two places and that the hyomandibula was not part of the jaw suspension. These features identify Helicoprion as a member of the stem holocephalan group Euchondrocephali. Our reconstruction illustrates novel adaptations, such as lateral cartilage to buttress the tooth whorl, which accommodated the unusual trait of continuous addition and retention of teeth in a predatory chondrichthyan. Helicoprion exemplifies the climax of stem holocephalan diversification and body size in Late Palaeozoic seas, a role dominated today by sharks and rays.”

FIgure 3. Closeup of figure 2.

FIgure 3. Closeup of figure 2.

The YouTube videos below
further emphasize the shark-like hypothesis and high-energy, fast-swimming method of cutting soft squid-like prey in half with the pieces falling to the sea floor. The videos don’t show the shark actually swallowing any prey. That is left to the imagination.

If instead, a chimaera body form is employed,
complimenting the authors’ statement that the specimen is a type of ratfish, then a low-energy, slow-swimming lifestyle should be inferred (using phylogenetic bracketing). Fossils are found in phosphate mines. In the present day deep cold waters carry three times as much phosphate as do warmer surface waters. Beyond those limits phosphate can precipitate out to form sea sediment in all phases and temperatures of phosphate solutions. That includes the weathering of terrestrial phosphatic rocks into nearby shallows.

Helicoprion teeth rarely show wear,
which is otherwise a good reason for getting rid of old, dull teeth in most vertebrates. Helicoprion throats may be tall, but they are also extremely narrow, AND blocked by the large median tooth whorl. That makes eating large prey difficult. Finally, Helicoprion grew to be giants that swam over phosphatic sea floors. Given these parameters and limitations,  what did Helicoprion swallow and how did it subdue prey?

According to Wikipedia
“The spotted ratfish swims slowly above the seafloor in search of food. Location of food is done by smell. Their usual hunting period is at night, when they move to shallow water to feed.” Ratfish feed on crabs and clams, along with shrimp, worms, small fish, small crustaceans, and sea stars, all bottom-dwelling prey.

What was available to eat back then
in sufficient quantities to sustain Helicoprion and never break off any teeth? Well, as it happens the largest prey items on the Permian seafloor (Fig. 5) were also the softest, slowest, most plentiful and easiest to find and graze on at night for a growing Helicoprion: the tall sponges. How Helicoprion fed (= what technique it used) on those sponges remains something to be imagined at present. Did Helicoprion nibble from the top of each sponge stalk? That’s my guess. If so, a central tooth whorl might have worked to break up each sponge stalk like a pizza cutter. Thereafter the mouth and gills worked together to suck in the broken sponge pieces.

Figure 5. Permian sea floor at night. The largest prey items here are also the softest and most plentiful and easiest to graze on at night for a growing Helicoprion: the tall sponges. Just guessing giving the present data.

Figure 5. Permian sea floor at night. The largest prey items here are also the softest, slowest, most plentiful and easiest to find and graze on at night for a growing Helicoprion: the tall sponges. Just guessing giving the present data.

Final note on sponges and precipitating phosphates
Colman 2015 reports, “The authors present strong evidence for polyphosphate (poly-P) production and storage by sponge endosymbionts. Zhang et al. also may have detected apatite, a calcium phosphate mineral, in sponge tissue. This work has major implications for our understanding of nutrient cycling in reef environments, the roles played by microbial endosymbiont communities in general, and aspects of P cycling on geologic timescales.”

Figure 2. Onychodus and Ischnacanthus share enough traits to make them sisters, apart from Brachyacanthus + Pteronisculus.

Figure 4. Onychodus and Ischnacanthus share enough traits to make them sisters, apart from Brachyacanthus + Pteronisculus.

For comparison
and by convergence two other fish have a median tooth whorl, though much smaller and much more conservative in both cases: Onychodus and Ischnacanthus (Fig. 4). So tooth whorls are in the gene pool, though rarely expressed.


References
Bendix-Almgreen SE 1966. New investigations on Helicoprion from the Phosphoria Formation of south-east Idaho, U.S.A. Biologiske Skrifter udgivet af det Kongelige Danske Videnskabernes Selskab, 14:1–54.
|Coleman AS 2015. Sponge symbionts and the marine P cycle. PNAS 112(14):4191-–4192.
Karpinsky AP 1899. On the edestid remains and its new genus Helicoprion. Zapiski Imperatorskoy Akademii Nauk, 7:1–67. (In Russian)
Lebedev O 2009. A new specimen of Helicoprion Karpinsky, 1899 from Kazakhstanian Cisurals and a new reconstruction of its tooth whorl position and function. Acta Zoologica, 90:171–182.
Mutter RJ and Neuman AG 2008a. New eugeneodontid sharks from the Lower Triassic Sulphur Mountain Formation of Western Canada. Geological Society, London, Special Publications 295:9–41.
Mutter RJ and Neuman AG 2008b. Jaws and dentition in an Early Triassic, 3-dimensionally preserved eugeneodontid skull (Chondrichthyes) Acta Geologica Polonica, 58 (2), 223-227.
Purdy RW 2008. The Orthodonty of *Helicoprion. *http://paleobiology.si.edu/helicoprion/
Zhang et al. 2015. Phosphorus sequestration in the form of phosphate by microbial symbionts in marine sponges. PNAS 112(14):4381–4386.

https://www.nationalgeographic.com/science/phenomena/2014/09/03/bizarre-prehistoric-ratfish-chomped-prey-with-buzzsaw-jaws/

Brian Switek writing in Wired.com 2011.

wiki/Helicoprion

SVP abstracts – Silurian gnathostome

Zhu et a. 2019 bring us
a new Silurian fish they claim is close to the origin of jawed vertebrates (= Gnathostomata).

From the abstract:
“Modern jawed vertebrates or crown-group gnathostome include the last common ancestor of living bony and cartilaginous fishes and all its descendants. The gross morphology of the earliest modern jawed vertebrates, and how they arose from stem gnathostomes, were previously unknown due to a lack of articulated fossils.”

These taxa are not unknown in the large reptile tree (LRT, 1592 taxa). Put enough taxa in an analysis and one will end up close to the origin of gnathostomes. There will be a last common ancestor. In the LRT Thelodus, a ?jawless (phylogenetic bracketing indicates some sort of transverse jaws are present) Silurian fish is the current proximal outgroup to all tested taxa with jaws. In LRT the extant whale shark (Rhincodon), angel shark (Squatina) and horn shark (Heterodontus) are basal members of the Gnathostomata and the first taxa with primitive tooth carpets.

“The recent discovery of the Xiaoxiang Fauna from the Silurian of South China revolutionarily adds to the diversity of Silurian jawed vertebrates. However, considerable morphological gap is still present between stem- and crown-group gnathostomes.”

Not so, when appropriate taxa are included.

“Here, we report a new bony fish very close to the crown-group gnathostome node, also from the Xiaoxiang Fauna. The attributed specimens include a head, jaws and an articulated postcranial skeleton.”

“The new fish displays a unique suite of characters: the dermal pectoral girdle condition transitional between Entelognathus and osteichthyans, the braincase profile recalling the condition in Janusiscus and early chondrichthyans, and the premaxillae and lower jaw largely showing osteichthyan features. This mosaic character combination suggests the tentative phylogenetic position of this new taxa in the most basal segment of the osteichthyan stem, possibly forming a quintessential component of the evolutionary transition between placoderms and osteichthyans.”

In the LRT taxa between placoderms and osteichthyans are either acanthodians (spiny sharks) on one branch, or catfish (also with spiny fins) on the other branch. Catfish are whales-shark mimics with regard to their jaws and teeth, likely representing some sort of reversal to that basal condition.

“For the first time, we are able to look into a near-complete bony fish close to the last common ancestor of all the living jawed vertebrates, and reconstruct the acquisition sequence of osteichthyan characters based on a series of fossils in morphological proximity. The fact that most of these fossils are from the Silurian Xiaoxiang Fauna, suggests that this fauna is unprecedentedly close to the initial radiation of jawed vertebrates.” 

Figure 2. Updated subset of the LRT focusing on basal vertebrates (fish). Arrow points to Hybodus. This tree does not agree with previous fish tree topologies.

Figure 1. Updated subset of the LRT focusing on basal vertebrates (fish). Arrow points to Hybodus. This tree does not agree with previous fish tree topologies. Check out the LRT for a slightly updated version of this cladogram.

This is all very interesting, and welcome, but let them look at the structure of Rhincodon as it relates to Thelodus at least once before settling down with the Zhu et al. hypothesis.

Figure 4. Manta compared to Thelodus and Rhincodon. All three have a terminal mouth essentially straight across, between the lateral eyes.

Figure 4. Manta compared to Thelodus and Rhincodon. All three have a terminal mouth essentially straight across, between the lateral eyes.


References
Zhu et al. 2019. A new Silurian bony fish close to the common ancestor of crown gnathostomes.