The deep history of the Hagfish.

The deep oceans serve as refugia for many groups of Animals which have effectively vanished in the shallow seas, including Coelacanths, Vampire Squid, Crinoids and Brittlestars, the living graptolite genus Rhabdopleura and other colonial Hemichordates, and several lineages of deep-sea Isopods, all of which seem to have diverged from their closest shallow-marine relatives more than 200 million years ago. Although generally thought of as evolutionary relicts, most of these groups appear to have undergone significant evolutionary diversification since entering the deep seas.

The Vertebrates underwent their first major evolutionary radiation in the oceans between the Ordovician and the Devonian, or between about 480 and 360 million years ago. Most living deep-sea Vertebrates, however, belong to a few relatively young groups, stemming from diversification events less than 100 million years ago. 

Although the first Vertebrates were jawless, by the End of the Devonian these had largely been eclipsed by jawed taxa, and today only two groups of Jawless Vertebrates survive, the Hagfish, Myxiniformes, and the Lampreys, Petromyzontiformes. The relationship between these groups, as well as the timing of divergence events between them, and within each group, remains unclear, though comparative genomic analysis has now confirmed that the two groups can be regarded as sister taxa (a variety of other relationships had been proposed, including a sister relationship between the Lampreys and Jawed Vertebrates, with the Hagfish being more distantly related to the two).

Hagfish form a significant proportion of the total biomass of Vertebrates on the deep ocean floor, with most species found on the continental slopes and ocean floors, between 200 m and 3 km beneath the surface, where they form an important part of the benthic ecosystem. Some species are found on the shallower ocean shelves, but these are rare. 

Hagfish have an Eel-like body with poorly developed eyes, a loose, scale-less skin, a minimal skeleton comprising a cartilaginous skull and rudimentary vertebrae, several auxiliary hearts, a mouth surrounded by barbels (short tentacles), a single nostril, and a single semicircular canal. The tongue of the Hagfish comprises a cartilaginous plate with two pairs of horny teeth, used to seize food and draw it into the mouth. 

In a paper published in the journal BMC Ecology & Evolution on 13 June 2024, Chase Doran Brownstein of the Department of Ecology and Evolutionary Biology at Yale University, and Thomas Near, also of the Department of Ecology and Evolutionary Biology at Yale University, and of the Yale Peabody Museum, present a time-calibrated phylogenetic tree for the Myxiniformes using data from fossils as well as a genetic dataset which includes 60% of living species of Hagfish.

Brownstein and Near were able to obtain sequences for the mitochondrial COI and 16S ribosomal DNA genes for 44 species of Hagfish from the GenBank database. This sample included two species of Rubicundus, two species of Neomyxine, 14 species of Myxine, and 26 species of Eptatretus, with an additional three potential species of Eptatretus from India, Japan, and Korea. the problematic ‘Notomyxine’ (= Myxine) tridentiger and several species   previously classified in ‘Quadratus’ and ‘Paramyxine'. 

This represents more than 50% of all known extant Hagfish species, although it does not include the problematic genus Nemamyxine, known only from two preserved specimens collected in the mid-twentieth century, with no genetic material available. This makes it impossible to place the genus Nemamyxine within a phylogenetic tree based upon genetic analysis, although Brownstein and Near note that it is thought to have close affinities to the genus Rubicundus, but also that there are problems with the validity of the genus. Both Nemamyxine and Rubicundus are defined as having an extremely slender body and an anteriorly placed ventral finfold that originates anterior of the ventral gill apertures, but this is also seen in many members of the genera Myxine and Eptatretus, as well as the Late Cretaceous fossil Hagfish, Tethymyxine tapirostrum. Nemamyxine is also defined as having a slender body depth and high slime pore counts, but these are also widespread in elongated Hagfish.

The first of two known specimens of Nemamyxine elongata, one of two described species in the genus Nemamyxine, which was found dead in a net in the Kaituna River on the Bay of Plenty on North Island, New Zealand, in 1958, and thought to have been a fishery discard. A second specimen was later recovered by a trawler from  the Canterbury Bight on the east coast of South Island, from a depth of between 132 and 140 m. Museum of New Zealand Te Papa Tongarewa.

Brownstein and Near constructed phylogenies using both maximum likelihood and Bayesian methods, and the online Clustal Omega tool at the European Molecular Biology Laboratory - European Bioinformatics Institute online resource portal to aid in 16S alignments. For outgroups they used the jawed Ornate Birchir, Polypterus ornatipinnis, West African Lungfish, Protopterus annectens, and Australian Ghostshark, Callorhincus milii, and the Lampreys Geotria australis, Petromyzon marinus, and Lampetra fluviatilis.

A molecular clock methodology with fossils was used to calibrate the divergence of clades. This is challenging for Hagfish, as the fossil record for the group is extremely limited, and most fossils assigned to the group are poorly preserved and/or of dubious placement. The putative stem-hagfish Myxinikela siroka from the Late Carboniferous Francis Creek Shale of Illinois was included in the study, as was the Late Cretaceous crown group Hagfish Tethymyxine tapirostrum from the Hâdjula Lagerstätte of Lebanon, as were a number of fossil Lampreys (phylogenetically the closest group to the Hagfish).

The putative stem-Hagfish Myxinikela siroka from the Late Carboniferous Francis Creek Shale of Illinois. Miyashita (2020).

Data on the habitat preference of Hagfish species was collected from the FishBase database, with two identified environments, continental shelf (less than 200 m) and continental slope (more than 200 m). These were used to make a probability-based estimation of the ancestral state of Hagfish groups using the R package, with the possibility that species might be flexible in their choice of habitat taken into account using the fitpolyMk function.

Brownstein and Near consistently recovered the three major lineages of Hagfish (the Rubicundinae, Eptatretinae and Myxininae) as  valid and distinct taxa. The genus Neomyxine was recovered as the sister taxon the genus Myxine within the family Myxininae,  rather than being the sister group to all other extant Hagfish, as have been found by some previous studies. The Family Rubicundinae was recovered as the outgroup to other extant Hagfish, something which has been found by some previous studies. The study also suggests that the genera Quadratus and Paramyxine should be included within the genus Eptatretus, and the species Notomyxine tridentiger should be included within the genus Myxine.

Hagfish phylogeny and tempo of diversification. Tip-dated Bayesian maximum clade credibility phylogeny of jawless fishes from two independent runs in BEAST 2.6.6 showing the interrelationships of the major lineages of hagfishes. Bars indicate 95% highest posterior density intervals for divergence times at nodes. Outgroups not shown. Grey bars are at nodes supported by posterior values of 0.90 or more, clear bars are at nodes supported by posterior values of 0.89 or less. Gray columns indicate mass extinction events. Dagger (†) indicates extinct species known from the fossil record. Pie charts indicate ancestral state reconstructions of habitat for each node, where purple represents the probability of a slope component (either slope or shelf-slope) at each node and salmon indicates the probability of continental shelf habitat being ancestral. Inset includes the transition matrix from the polymorphic character ancestral reconstruction analysis (note that purple here is exclusively slope, as opposed to purple denoting slope/shelf-slope at nodes in the phylogeny). Photograph of Eptatretus stoutii is courtesy Douglas Fudge. Brownstein & Near (2024).

The results of the study suggest that the three major Hagfish groups diverged from one-another during the Palaeozoic. This did not change when the Carboniferous Myxinikela siroka was included in the matrix, suggesting that the use of this taxon as a calibration point is valid. Brownstein and Near note that they excluded the Mazon Creek 'Hagfish' Gilpichthys greenei from the study, as the affinities of this abundant fossil are now considered highly doubtful. Other phylogenetic studies have included this species, recovering it as either a stem Hagfish, or a Jawless Fish of uncertain affinities. Brownstein and Near suggest that these fossils may be difficult to interpret phylogenetically as most had decayed somewhat before preservation.

The putative Hagfish Gilpichthys greenei from the Mazon Creek fossil beds. Earth Science Club of Northern Illinois.

Brownstein and Near consistently found that the crown Hagfish (a crown Hagfish is any species, living or fossil, which is descended from the last common ancestor of all living Hagfish) arose in the Early Permian, and the split between the Eptatretinae and Myxininae occurred in the Early Triassic. Both events are substantially older than previous studies have suggested, with the diversification of major Hagfish clades until now assumed to have happened in the Middle-to-Late Cretaceous. Brownstein and Near note that the use of mitochondrial DNA has been linked to the overestimation of the age of some groups of Ray-finned Fish, but cannot see how this would lead to the discrepancy between their study and earlier studies of Hagfish which also used mitochondrial DNA. Instead they suggest that the variance is due to the increased number of living species in their study, combined with a stricter approach to the inclusion of fossil species, with less certain species such as Gilpichthys greenei excluded. 

This revised timeline removes a 120 million year gap between the separation of the Hagfish and their closest relatives (the Lampreys), as well as showing that the group have survived three major extinction events, including the End Permian, which wiped out 81% of marine species. This makes the crown group Hagfish one of the oldest known Vertebrate crown groups, and far older than most other marine Vertebrate groups. 

The reconstruction of the Ancestral habits of the Hagfish suggests that the oldest members of the group occupied the continental slopes (more than 200 m beneath the surface) during the Late Palaeozoic. This is despite all known fossil Hagfish coming from coastal slope or estuarine environments. All the major Hagfish groups apparently first appeared on the continental slopes, or at least as organisms with flexible requirements able to inhabit both the continental slopes and shelves.

Hagfish and Lampreys have been the sole surviving jawless Vertebrates since the Triassic Extinction. This makes them important to our understanding of the earliest Vertebrates, although probably atypical of these. 

Brownstein and Near's study suggests that the crown group Hagfish emerged during the Permian, with the three major extant groups having appeared by the end of the Early Triassic, 20-30 million years after the oldest putative Hagfish fossils. It is likely that the stem group Hagfish appeared during a significant radiation event after the extinction of the jawless Ostracoderms at the end of the Devonian. 

Hagfish have a simple bodyplan, which has remained essentially unchanged for a very long time, notably so compared to other ancient Vertebrate groups such as the Teleosts, Chondrichthyans, and Lissamphibians. This highly specialised anatomy appears to have developed before the End of the Permian.

This deep diversification is different to the situation seen in Lampreys, where the extant groups all appear to have derived from a series of regional diversification events within the past 100 million years. Hagfish species appear to have diverged from their closest relatives an average of 31.6 million years ago, compared to 1-2 million years for most Lampreys. The most ancient division for a single species is that for Eptatretus cheni, which appears to have diverged from other members of the genus Eptatretus in the Jurassic. This is a similar timing for the division between the living Neoselachian Sharks and Rays, the Tuatara, Sphenodon punctatus, and the Squamates, or the Salamanderfish, Lepidogalaxias salamandroides, and all other Teleosts.

Eptatretus cheni, not notably different to other members of the genus Eptatretus, but separated from them since the Jurassic. Fish Database of Taiwan/FishBase.

Hagfish taxonomy is a challenging field, due to the conservative morphology of these organisms, and the inaccessible environments in which they live. The widespread genus Rubicundus is the only genus in the family Rubicundinae, and forms the sister group to all other Hagfish, but was not recognised as a distinct genus until 2013. Brownstein and Near's study implies that this genus split from its closest living relatives in the Permian. 

A Pink Hagfish, Rubicundus eos. The genus Rubicundus appears to have diverged from all other extant Hagfish in the Permian. Museum of New Zealand Te Papa Tongarewa.

Brownstein and Near's study also highlights that deep marine habitats have been utilised by Hagfish since the origin of the group in the Permian. Lineages of Myxine and Eptatretus found in shallower continental shelf environments appear to have diversified into these shallower waters relatively recently, with fossil Hagfish from shallow marine environments probably the result of similar diversification events. This makes the Hagfish the Vertebrates the group with the longest history in deep marine environments, with a continuous habitation of these environments long predating the arrival of the ancestors of any extanct Chondrichthyan or Teleost found in the deep seas. 

This inhabiting of deep-sea environments may explain how the group has persisted so long with relatively little apparent evolutionary innovation. Although the group has not  occupied deep marine environments for as long, the oldest surviving Chondrichthyan lineages, such as the Goblin Sharks, Frilled and Sevengill Sharks, Chimeras, and Ratfish, all inhabit deep environments. Thus thee deep sea. environment appears to be a refugia for Vertebrate groups able to live there, offering a degree of protection against  extinction events which heavily impact the shallow seas.

Nevertheless, Hagfish appear to have undergone significant diversification within deep sea environments, with many distinct lineages arising over the time they have dwelt there.

Most Vertebrate groups found in the deep seas have colonised these environments within the last 100 million years. In contrast, many Invertebrate groups have long deep marine lineages. This has led to the view that the deep seas can act as a refugia for groups that can live there during mass extinction events that affect the shallow seas.  Until now, no Vertebrate group has been seen as truly endemic to this refugium, but Brownstein and Near's study suggests that the deep seas are the principle habitat for Hagfish, with modern and fossil shallow-water species being the result of repeated colonisations from deeper marine environments.

See also...

A new specimen of the Hupehsuchian Nanchangosaurus throws light on the origin of the earliest Mesozoic Marine Reptiles.

The first groups of Mesozoic Marine Reptiles returned to the seas early in the Triassic, quickly gaining suites of specialist adaptations to pelagic life which have made it hard to determine their closest terrestrial relatives. The most notable of these groups is the Ichthyopterygia (Ichthyosaurs and close relatives), a group which for a long time were considered to be of uncertain affinities, and which today are placed within the Diapsida, but with no widely accepted hypothesis on their relationship to other members of this group. The sister group to the Ichthyopterygia is considered to be the Omphalosauridae, another group of highly modified pelagic Marine Reptiles, with the two groups forming the Ichthyosauriformes.

The Hupehsuchians are a group of small Mesozoic Marine Reptiles known only from the Early Triassic deposits of the Nanzhang-Yuan’an region of Hubei Province, South China. As with other Marine Reptile groups, their affinities were for a long time hard to determine, though in the last decade it has been realised that they are the sister group to the Ichthyosauriforms, with the two groups together being referred to as the Ichthyosauromorphs. Hupehsuchians had elongate bodies with slender heads, dermal plates on their dorsal surfaces, dense ribs and tightly packed gastralia. Importantly, they were fairly small, ranging from 40 cm to about 2.3 m in length, and restricted to shallow-marine environments, suggesting that they may be closer to the terrestrial origins of the Ichthyosauromorphs than other members of the group.

A large number of Hupehsuchian specimens have been found, and the anatomy of the group reasonably well understood. However, almost all of these specimens are preserved in lateral view, with only three specimens known with their skulls preserved in ventral view, which presents challenges when comparing Hupehsuchians to other groups

In a paper published in the journal Historical Biology on 25 May 2024, Jun Liu, Fan Wu, and Yu Qiao of the Division of Geology at Hefei University of Technology describe a new Hupehsuchian specimen from the Early Triassic Jialingjiang Formation in Yuan’an County, in the west of Hubei Province, China, which is preserved in ventral view, and discuss the implications of this for our understanding of the origins of the Hupehsuchians and related groups.

The specimen, HFUT YAV-10-001, is a small, well-preserved Hupehsuchian on a slab and counter-slab in ventral view, and held in the collection of the Geological Museum of Hefei University of Technology. This comprises the skull, a set of 36 articulated vertebrae and associated ribs, and a partial appendicular skeleton, including both pectoral girdle, forelimbs and partial hindlimbs. The majority of the bones are preserved on the main slab, while vertebral and rib fragments and the right pectoral girdle and forelimb elements are preserved on the counter slab. 

The Hupehsuchian HFUT YAV-10-001 from the Early Triassic of Hubei Province, South China. (A) Photograph of the skeleton on the main slab. (B) Photograph of the skeleton on the counter slab. (C) Interpretive drawing of the skeleton on the main slab. (D) Interpretive drawing of the skeleton on the counter slab. Scale bars are 1 cm. Abbreviations: 2 and 4, distal carpals; I, IV and V, metacarpals; ax, axis; axr, axis rib; bo, basioccipital; Cl, clavicle; cna#, cervical neural arch; Co, coracoid; da, dermal armour; dna#, dorsal neural arch; dns#, dorsal neural spine; dnss, dorsal neural spine second (distal) segment; dr#, dorsal rib; F, femur; Fi, fibula; H, humerus; i, intermedium; mand, mandibular ramus; pm, premaxilla; R, radius; r, radiale; Sc, scapula; Ti, tibia; U, ulna; u, ulnare. Jun et al. (2024).

Jun et al. consider HFUT YAV-10-001 to be referable to the genus Nanchangosaurus, having the same number of cervical vertebrae, as well as distinctively low neural spines, as well as lacking a parapophysis, and having subequally-sized scapula and coracoid bones. However, it differs from the previously described Nanchangosaurus suni in a number of features, including having forelimbs which are longer compared to the size of the Animal, and a smaller overall size, with Nanchangosaurus suni reaching about 20 cm in length, while HFUT YAV-10-001, which is clearly an adult, has an estimated size of only 15 cm. This could be a sign that HFUT YAV-10-001 represents a new species of Nanchangosaurus, but it may also indicate that Nanchangosaurus suni was sexually dimorphic, something which has been suggested in early Ichthyosauriforms and Sauropterygians, and which might therefore be predicted in an early Hupehsuchian. For this reason, Jun et al. assign specimen HFUT YAV-10-001 to Nanchangosaurus cf. suni.

The different aspect in which HFUT YAV-10-001 enables significant extra features to be added to a matrix used for the phylogenetic analysis of Diapsidans. Jun et al. recover HFUT YAV-10-001 as a Hupehsuchian, and the Hupehsuchians as the sister group to the Ichthyosauriformes, together  forming the Ichthyosauromorphs, as with previous studies. Their analysis further suggests that the Ichthyosauromorphs form the sister to the Sauropterygomorpha (a diverse group of Mesozoic Marine Reptiles which included groups such as the Nothosaurs and Plesiosaurs), and that this larger grouping forms a sister group to the Thallatosauria, a group of Lizard-like Marine and semi-Marine Reptiles, again restricted to the Triassic.  This grouping of the Ichthyosauromorphs, Sauropterygomorphs, and Thalattosaurs was in turn found to be the sister group to the Archosauromorphs, the group that includes the Archosaurs (Pterosaurs, Crocodilians, and Dinosaurs) plus close outgroups such as the Rhynchosaurs and Tanystrophids.

Simplified phylogeny of Sauria showing the relationships of Ichthyosauromorphs to other Reptiles. Jun et al. (2024).

Sea also...

Kermitops gratus: A new species of Amphibamiform Temnospondyl from the Early Permian Lower Clear Fork Formation of Texas.

Modern Lissamphibians (Frogs, Salamanders, and Caecillans) are considered to be descended from a group of small, lightly-built, terrestrial Temnospondyls called the Amphibamiforms, known from Carboniferous, Permian, and Triassic deposits. The assumed relationship between Amphibamiforms and Lissamphibians is based upon the presence of bicuspid, pedicellate teeth in some adult Amphibamiforms, one of few traits which is common to all modern Lisamphibian groups. The general skull shape of Amphibamiforms also tends to resemble that of Lissamphibians, though these, as with the skeletons in general, are more simplified in Lissamphibians than in Amphibamiforms, presumably as a result of changes in timing and rate of developmental processes in Lisamphibians.

In a paper published in the Zoological Journal of the Linnean Society on 21 March 2024, Calvin So of the Department of Biological Sciences at George Washington University, Jason Pardo of the Negaunee Integrative Research Center at the Field Museum of Natural History, and Arjan Mann, also of the Integrative Research Center at the Field Museum of Natural History, and of the Department of Paleobiology at the Smithsonian National Museum of Natural History, describe a new species of Amphibamiform Temnospondyl from the Early Permian Lower Clear Fork Formation of Texas.

The new species is described from a single partial skull comptising a near complete roof and  occiput with a partial braincase, and mandibles. It is given the name Kermitops gratus, where 'Kermitops' is a combination of the name 'Kermit', in reference to the famous Lissamphibian and beloved Muppets’ character created and originally performed by Jim Henson, and '-ops', the Greek for 'face', while 'gratus' means 'gratitude' in Latin, in thanks to the Smithsonian National Museum of Natural History vertebrate palaeontology curator Nicholas Hoton III, and other members of the Smithsonian National Museum of Natural History field party that were involved in the collection efort.

Photograph (A) and interpretive illustration (B) of Kermitops gratus (USNM PAL 407585) in dorsal view. So et al. (2024).

The skull is approximately 3 cm long along the midline and 2 cm wide at the level of the occiput. There is some taphonomic distortion on both sides, making the orbits appear slightly more ovoid than they would have in life, and the left orbit is partially disarticulated. The anterior palate and braincase are lost but the remainder of the skull is well-preserved, even showing a full arrangement of palpebral ossicles in place and showcasing fine dermal ornamentation on the dorsal skull. The margins of the orbit are slightly raised, resulting in a differentiation of the orbital margin from the rest of the skull roof surface. The snout is long and parabolic in shape, consistent with the morphology seen in Micropholids. 

Attempts to include Kermitops gratus in phylogenetic trees using different methods produced quite different results, suggesting that insufficient sampling of the group has been done to achieve a consensus hypothesis. Notably, So et al. failed to find a clear relationship between Lissamphibians and Amphibamiform taxa with pedicellate bicuspid teeth, which would seem to indicate either that the trait evolved separately numerous times within the Amphibamiformes, or that it was present in the earliest members of the group, and lost multiple times.

See also...

Timorebestia koprii: A giant stem-group Chaetognath from the Early Cambrian Sirius Passet Lagerstätte of Peary Land, North Greenland.

The Cambrian Explosion was marked by a sudden radiation of numerous Animal taxa, and the simultaneous expansion into numerous ecological roles. Exactly why this started remains unclear, but the evolution of predatory behaviour and the subsequent arms race between predators and prey is considered to be a significant part of the subsequent rapid evolutionary divergence. The 'explosion' is now recognised to have had three distinct steps; the appearance of Worm-like organisms in the terminal Ediacaran, the appearance of hard parts in the earliest Cambrian, and an rapid evolutionary diversification which produced the majority of modern phyla in Cambrian Stage 3.

Chaetognaths, or Arrow Worms, may have been the first group of Bilaterian Animals to specialise in hunting within the water column, with the Paraconadonts of the Earliest Cambrian now recognised as the grasping spines of early Chaetognaths. Amiskwia sagittiformis, a Burgess Shale fossil first described by Charles Doolittle Walcott in 1911, shows a strong similarity to modern Chaetognaths, possessing a roughly tubular body with paired lateral and tail fins. Despite this, it was for a long time thought unlikely to be a Chaetognath, as no specimen had ever been found with grasping spines, considered to be a key feature of the group. However, recent studies of these fossils have shown that they have an internal jaw apparatus, similar to that seen in Gnathiferans, a group which phylogenetic studies have suggested are closely related to the Chaetognaths, with the two forming a single clade, the Chaetognathifera. This combination of Chaetognath-like paired lateral and tail fins with an Gnathiferan-like internal jaw apparatus, raises interesting questions about the origin of these groups; where the two traits present in the ancestors of all Chaetognathiferans? Or was Amiskwia sagittiformis a member of one lineage that had had convergently gained a trait associated with the other? Understanding the answer to this question could have implications for our understanding of not just the origins of the four modern phyla, the Chaetognatha, Gnathostomulida, Micrognathozoa, and Rotifera.

In a paper published in the journal Science Advances on 3 January 2023, a team of scientists led by Tae-Yoon Park of the Division of Earth Sciences at the Korea Polar Research Institute, and the University of Science and Technology, describe a new, giant, stem-group Chaetognath from the Early Cambrian Sirius Passet Lagerstätte of Peary Land, North Greenland, and discuss its implications for the origin of the group.

The new species is named Timorebestia koprii, where 'Timorebestia' means 'fear-inducing beast' (it also appears to be a reference to Robert Burn's line 'timorous beastie', from the poem 'To a Mouse', though this is not stated), and 'koprii' derives from KOPRI, the acronym for the Korea Polar Research Institute. The species is descibed from 13 specimens collected from several horizons at the main Sirius Passet Lagerstätte locality; specimens were found over 12 m of exposure, although most were from within a particular fossiliferous interval of between 5 m and 7 m.

Holotype (MGUH 34286) of Timorebestia koprii. (A) to (C) Entire specimen. (D) and (E) Jaw apparatus in the anterior region of trunk. (A) Wavelength-dispersive x-ray spectrometry map of carbon on the specimen surface. (B) Polynomial texture mapping visualization using specular enhancement, illuminated from top left. (C) Interpretative drawing. (D) Carbon map of jaw apparatus indicating some indistinct enrichment of carbon within it. (E) Polynomial texture mapping image illuminated from top left of jaw apparatus. (F) Interpretative drawing of jaw apparatus based on tracing of multiple illumination angles. Abbreviations: Bp, basal plate; Lb, lateral bars; Jw, jaw; Ja, jaw apparatus; G, gut; Tm, transverse muscles; Fr, fin rays; Ps, posterior structure; Lm, longitudinal muscles. Park et al. (2024).

Timorebestia koprii is a wide bodied 'Amiskwiiform' with lateral fins running along the majority of the length of its trunk and a well-developed rounded caudal fin; these fins are rayed, with no division between the trunk and caudal fins. It's fore-end has a distinct tapering 'head' with paired antennae. An internal jaw apparatus is preserved as a pair of blunt anterior elements, and a single anterior plate. longitudinal bands of discrete muscles can be seen in the trunk, as well as additional outer circular or transverse muscles, although these are much more sparse. A digestive tract can be seen running from the head to just in front of the caudal fin.

Additional specimens of Timorebestia koprii (A) MGUH 34287, the largest preserved individual imaged with high dynamic range based on multiple images taken with different incident illumination angles while submerged in water. (B) Interpretative drawing. (C) MGUH 34288, another very large individual preserving less detail in low angle illumination. (D) High dynamic range image. (E) Interpretative drawing. (F) MGUH 34289, laterally preserved specimen. (G) Interpretative drawing. Abbreviations: An, antennae; Cr, caudal region; Hd, head; Mu, muscles; Mgc, mineralized gut contents; G, gut; Tr, trunk. Park et al. (2024).

Timorebestia koprii shows considerable variation in size, with the smallest being about 22 mm in length, while the largest is an incomplete specimen with a preserved bodylength of 206 mm, plus 92 mm antennae, giving a total preserved length of 298 mm, and an estimated original length in excess of over 300 mm. This is remarkable, as most extant Chaetognaths are only a few millimetres in length, while the largest living species, Pseudosagitta gazellae, reaches about 10 cm in length, similar to the largest described fossil species to date, Capinatator praetermissus, from the Burgess Shale of British Columbia.

Digital 3D model of Timorebestia koprii. Reconstruction showing internal and external anatomy (red, musculature; blue, ventral ganglion; black, jaw apparatus; green, gut). (A) Lateral view. (B) Dorsal view. (C) Ventral view. (D) Ventral view excluding musculature. Park et al. (2024).

Timorebestia koprii shares a general bodyplan with Amiskwia sagittiformis, which is not seen in any extant group of Gnathiferans. Since Amiskwia sagittiformis was the first discovered organism with this bodyplan, Park et al. refer to these Animals as 'Amiskwiiforms'. They also note the presence of another, as yet unnamed stem-Chaetognath in the Sirius Passet Lagerstätte, which does have external grasping spines, and is interpreted as being closer to the crown group than either Timorebestia koprii or Amiskwia sagittiformis. Thus it is assumed that an internal jaw apparatus is the ancestral state in Chaetognaths and Gnathiferans. The accessory pair of transverse elements within these internal jaws resembles the uncus elements of Rotifers (which are closely related to Chaetognaths and Gnathiferans). The external transverse or circular muscles seen in Amiskwiiforms are absent in modern Chaetognaths, however, Rotifers and Mictognanozoans have strong circular muscles and Gnathostomulids have numerous, but much thinner circular muscles, similar to those seen in Timorebestia koprii, suggesting that this might be the ancestral trait in Chaetognathiferans. Both Timorebestia koprii and the unnamed Sirius Passet species show phosphatized structures which appear to be a ventral ganglion with lateral neuron somata, something which strongly suggests they are more closely related to Chaetognaths than to other groups. A phylogenetic analysis carried out by Park et al. suggests that Timorebestia koprii and Amiskwia sagittiformis are the earliest (known) branching taxa on the Chaetognath branch of the Chaetognathiferan tree.

Ventral ganglion comparisons and phylogenetic relationships. (A) and (B) Timorebestia koprii MGUH 34290 and an interpretative drawing highlighting the presence of a paired set of bilobed structures (arrowed), mineralized by phosphate interpreted as lateral neuron somata of a ventral ganglion. (C) and (D) A small undescribed Chaetognath from Sirius Passet, MGUH 34299, preserving the ventral ganglion (arrowed) as paired phosphatized structures. (E) to (G) Confocal laser scanning microscope images of the extant Chaetognath Sagitta sp. (E) Histochemical labelling of nuclei (blue) and α-tubulin (green). (F) Same view as in (E), with only immunolocalization nuclei (blue). (G) Magnified view of the ventral nerve centre and the lateral neuron somata (arrowed) enriched in nuclei (blue). (H) Summary of phylogenetic analysis placing Timorebestia koprii on the Chaetognath stem. Schematic reconstructions at the tips indicate relative association of the jaw apparatus, pedal ganglion in Rotifers, and ventral ganglion. Park et al. (2024).

The large size of Timorebestia koprii is surprising; the largest known specimen of Amiskwia sagittiformis is only about 35 mm long, while the previous largest known fossil Chaetognath, Capinatator praetermissus from the Burgess Shale reaches only about 100 mm, comparable to the size of the largest living species, Pseudosagitta gazellae. The large size of Timorebestia koprii makes it one of the largest Early Cambrian pelagic species, which in combination with its complex swimming apparatus and long antennae, would probably have made it a top predator in its environment, an interpretation which is supported by the presence of the Bivalved Arthropod Isoxys volucris within the digestive tract of many specimens.

Reconstruction of Timorebestia koprii in the pelagic ecosystem preserved in Sirius Passet. Other taxa shown in the foreground are Kiisortoqia, Siricaris, Kerygmachela, Pauloterminus, Kleptothule, and Isoxys. Further in the background are two Radiodonts: Tamisiocaris and an Amplectobeluid. Robert Nicholls in Park et al. (2024).

This is surprising, as Chaetognaths are close to the bottom of the food chain in modern oceans, feeding on tiny zooplankton and, in the case of the largest species, very small Fish. However, they are one of (if not the) earliest pelagic predatory groups to appear, with Paraconadonts such as Protohertzina appearing in the oldest Cambrian deposits, so it is perhaps unsurprising that they were for a time the top predators in the Cambrian oceans. The first Panarthropods are thought to have colonised the water column around the transition from Cambrian Stage 2 to Cambrian Stage 3, or roughly 525 to 522 million years ago. The Sirius Passet Lagerstätte is thought to be between 523 and 518 million years old, making it a window into a time when one ecological realm was replacing another, including the replacement of Chaetognaths as top pelagic predators by the emerging Panarthropods.

See also...

Online courses in Palaeontology. 

Follow Sciency Thoughts on Facebook.

Follow Sciency Thoughts on Twitter.

Assessing the impact of predation on birth size in marine Snakes.

Moving from one environment to another exposes organisms to new selective pressures on life histories, and where multiple organisms from related lineages make the same transition, it presents an opportunity for biologists to analyse these pressures. For example, Squamates (Snakes and Lizards) which migrate from warm environments to cooler ones frequently switch from egg-laying to bearing live young, while Birds making the same transition tend to produce smaller clutches of eggs.

Birth size is considered to be a fundamental life-history trait, subject to a number of evolutionary pressures. Where intraspecific competition is low, smaller birth sizes are often a result, as offspring survival is not dependent on size at birth. Conversely, larger birth size can be driven by a number of factors, for example a lack of small prey can drive up birth size in species where the young must hunt for themselves, as only larger neonates are able to capture enough prey to survive. 

The shift from a terrestrial habitat to a marine one presents organisms with a variety of different challenges, including thermal regimes, oxygen availability, light levels, ocean currents, types of predators, prey, competitors and pathogens. Nevertheless, the marine environment clearly presents opportunities for terrestrial Tetrapods, with numerous lineages of Mammals, Reptiles, and even Birds having made the transition. Elapid (Front-fanged) Snakes have made this transition at least three times, with the Sea Kraits, Laticaudinae, having split from terrestrial relatives in Asia about 16 million years ago, while at least two lineages within the Australian subfamily Hydrophiinae (together referred to as Sea Snakes) switched to a marine habit more recently. The three lineages show convergent evolution for a number of traits, including the development of laterally compressed bodies with paddle-like tails, the appearance of salt-excreting glands, and common life-history traits. A fourth group of (non-Elapid) Snakes, the Acrochordidae, are semi-aquatic, and often semi-marine in habit, and show some of these traits.

Marine Snakes typically produce fewer young than terrestrial Snakes, which has been linked to a need for gravid females to retain a hydrodynamic shape. Nevertheless, the offspring are typically larger at birth than those of their terrestrial relatives, which would seem to work against this.

In a paper published in the journal Royal Society Open Science on 13 December 2023, Richard Shine of the School of Natural Sciences at Macquarie University, Shai Meiri of the School of Zoology and Steinhardt Museum of Natural History at Tel-Aviv University, Terri Shine and Gregory Brown, also of the School of Natural Sciences at Macquarie University, and Claire Goiran of LabEx Corail and  Institut de sciences exactes et appliquées at the Université de la Nouvelle-Calédonie, examine the possibility that size-selective predation on young Snakes could be the driver of increased neonatal size in Marine Snakes.

Smaller terrestrial Snakes are known to be vulnerable to a wider range of predators than larger Snakes, with many predators targeting smaller Snakes while actively avoiding larger ones. However, predation rates on smaller Snakes can be lower than on larger individuals, due to the ability of small Snakes to remain inactive in well-hidden retreats.

Marine Snakes are less able to do this, as they must ascend to the surface to breath. This means that Snakes must leave their protective shelters and cross open water, where they are vulnerable to predation, several times per day. Predation of Snakes by large Fish during these crossings is well-documented, supporting the hypothesis that this is a risky endeavour for marine Snakes.

In order to test the hypothesis, Shine et al. first examined records of birth sizes in both marine and terrestrial Snakes, to confirm that the perceived trend was in fact real, then carried out experimental trials with model Snakes of different sizes to see if smaller Snakes were in fact more vulnerable to predation.

Shine et al. obtained data on hatchling and neonate sizes (Snakes can lay eggs or bear live young, but this does not appear to affect infant size much) and snout-vent lengths of adult females of 166 species of terrestrial, semi-aquatic, and marine Snakes, from published literature and the collection of the Steinhardt Museum. Semi-aquatic Snakes were found to produce slightly smaller offspring than terrestrial Snakes on average. However, the sample size for these Snakes was very small, and the subject was not investigated further. The adult snout-vent length for female Snakes in the study averaged at 800 mm, with the offspring of terrestrial Snakes having an average length of 200 mm, and the average length of new-born marine Snakes being 300 mm. 

Based upon this, Shine et al. hypothesised that a 200 mm Snake would be at significantly higher risk of predation in a typical marine Snake environment than a 300 mm Snake. To test this, an experiment was devised in which commercially available fibreglass fishing lures designed to resemble Snakes had their hooks removed and additional weights added to ensure they retained negative buoyancy, and were painted black to resemble the most common colour morph of the locally abundant Turtlehead Sea Snake, Emydocephalus  annulatus. These were then dragged by a snorkeler, Claire Goiran, over Coral reefs off the island of Ile aux Canards in New Caledonia, while a second snorkeler, Richard Shine, followed and recorded the reaction of large predatory Fish to the lures. 

A Camouflage Grouper, Epinephelus polyphekadion, following a black Snake-shaped lure, immediately prior to launching an attack. Teri Shine in Shine et al. (2023).

During 47 trials, Shine et al. recorded 114 responses. These included 38 attacks, and 76 encounters in which Fish followed the lure but did not attack. The size of the lure did not appear to influence whether or not Fish followed it, but they were significantly more likely to attack the smaller lures. Similarly, larger Fish were more likely to attack the lures, while smaller Fish tended to break off following without attacking. Thus, the majority of attacks were by large Fish on small lures.

Multiple lineages of Snakes which have invaded marine habitats have had an increase in neonatal size, combined with a reduced brood size (which are probably connected). Shine et al.'s study suggests that increased predation on smaller Snakes is a plausible explanation for this (although they stress that the results of their study cannot be taken as an absolute proof).

Shine et al. also note that larger Snakes are more likely to survive attacks by Fish, noting that two incidents of Snakes being seized by Fish and then released because the Fish was unable to overpower the Snake have been recorded on reefs close to their study area. In one of these incidents a Chocolate Grouper, Cephalopholis  boenak, unsuccessfully attacked a Turtlehead Sea Snake, Emydocephalus  annulatus, and in the other a Reef Stonefish, Synanceia verrucosa, was forced to break off an attack on a Blue Lipped Sea Krait, Laticauda  laticaudata, suggesting that larger size may present an advantage to young Snakes in surviving attacks, even if Fish do not discriminate against larger Snakes when choosing whether to attack.

A Reef Stonefish, Synanceia verrucosa, making an unsuccessful attack on a Blue Lipped Sea Krait, Laticauda  laticaudata, off the coast of Ile aux Canards in October 2022. Richard Fish/iNaturalist.

Predation is often cited as a likely cause of evolutionary pressure, influencing traits such as size and colouration. However, direct evidence of such impacts is difficult to gather accurate information on this unless predation rates are extremely high. Furthermore, it is difficult to design experiments looking at predatory behaviour for larger Animals without running into ethical and logistical constraints.

Predation is not the only driver of larger size in young marine Snakes which has been made, but it does seem to be the best supported by the available evidence. 

It has been suggested that larger size may provide an advantage when swimming, with smaller Snakes potentially being less efficient swimmers, using more energy to go slower. However, research into Sea Kraits has shown that smaller individuals have a higher swimming speed relative to crawling speed than larger individuals, suggesting that in these marine Snakes smaller size produces an advantage when swimming. 

Another possibility is that larger size in neonatal marine Snakes might be driven by prey size, with a shortage of suitable prey capturable by smaller Snakes creating a need for infant Snakes to be as large as possible. However, many Sea Snakes feed on smaller prey, notably members of the genus Emydocephalus are specialist feeders on Fish eggs, and several members of the genus Hydrophis have miniaturized heads and slender forebodies that enable them to penetrate the burrows of the small Fish upon which they prey.

Another possibility is that intraspecific competition drives larger size in young marine Snakes, with larger individuals excluding smaller individuals from better territories or access to prey. However, aggressive behaviour between members of the same species has never been observed in marine Snakes, making this unlikely.

Finally, larger size can act as a buffer against temperature changes, with larger bodies taking longer to either warm up or cool down that smaller bodies, thereby giving the Snakes more time to react to changes in conditions. However, marine environments offer much more protection against such temperature fluctuations than terrestrial ones, due to the high conductivity of water, making this highly unlikely as a driver of size in marine Snakes.

See also...

Follow Sciency Thoughts on Facebook.

Follow Sciency Thoughts on Twitter.