Skip to main content
Journal of Anatomy logoLink to Journal of Anatomy
. 2022 May 19;241(2):393–406. doi: 10.1111/joa.13689

Mechanisms of dermal bone repair after predatory attack in the giant stem‐group teleost Leedsichthys problematicus Woodward, 1889a (Pachycormiformes)

Zerina Johanson 1,, Jeff Liston 2,3,4, Donald Davesne 5, Tom Challands 6, Moya Meredith Smith 1,7
PMCID: PMC9296021  PMID: 35588137

Abstract

Leedsichthys problematicus is a suspension‐feeding member of the Mesozoic clade Pachycormiformes (stem‐group Teleostei), and the largest known ray‐finned fish (Actinopterygii). As in some larger fish, the skeleton is poorly ossified, but the caudal fin (tail) is well‐preserved. Bony calluses have been found here, on the dermal fin rays, and when sectioned, show evidence of bone repair in response to damage. As part of this repair, distinctive tissue changes are observed, including the deposition of woven bone onto broken bone fragments and the surface of the lepidotrichium, after resorption of the edges of these fragments and the lepidotrichial surface itself. Within the woven bone are many clear elongate spaces, consistent with their interpretation as bundles of unmineralized collagen (Sharpey's fibres). These normally provide attachment within dermal bones, and here attach new bone to old, particularly to resorbed surfaces, identified by scalloped reversal lines. Haversian systems are retained in the old bone, from which vasculature initially invaded the callus, hence bringing stem cells committed to forming bone onto the surfaces of the damaged area. These observations provide strong evidence of a vital response through survival of a predatory attack by a large marine reptile, coeval with Leedsichthys in the Jurassic seas.

Keywords: bone repair, dermal bone, Leedsichthys, lepidotrichia, osteocytes, stem‐teleost


Leedsichthys problematicus Woodward, 1889a is a suspension‐feeding member of the Pachycormiformes (stem‐group Teleostei), and the largest known ray‐finned fish (Actinopterygii). Bony calluses have been found on the dermal fin rays of the tail, and when sectioned, show evidence of bone repair in response to damage, including deposition of woven bone onto resorbed surfaces. These observations provide strong evidence of a vital response through survival of a predatory attack by a large marine reptile, coeval with Leedsichthys in the Jurassic seas.

graphic file with name JOA-241-393-g009.jpg

1. INTRODUCTION

With conservative size estimates of up to 16.5 metres in Standard Length (Liston et al., 2013), Leedsichthys problematicus Woodward, 1889a a member of the Mesozoic stem‐teleost group Pachycormiformes, is the largest known ray‐finned fish of all time (Actinopterygii; Ferrón et al., 2018). Pachycormiformes comprise several large suspension‐feeding ray‐finned fishes (Actinopterygii), including Leedsichthys, Asthenocormus, Martillichthys, Bonnerichthys and Rhinconichthys (Dobson et al., 2019; Friedman et al., 2010, 2013; Liston, 2008; Schumacher et al., 2016). These are Mesozoic ecological analogues to the modern basking shark, whale shark and baleen whales. The skeleton of Leedsichthys is poorly known, likely due to reduced ossification (Liston et al., 2013; Liston & Noè, 2004; Woodward,  1889b). Nevertheless, elements of the skull and gill arch skeleton have been described (Liston, 2016), along with the pectoral and dorsal fins, and in particular, the caudal fin (Liston, 2004, 2007; Liston et al., 2013).

The caudal fin of L. problematicus is primarily known from specimen NHMUK PV P10000, having been described as including large dorsal and ventral lobes, approximately equal in size (Liston, 2007; Liston et al., 2013: fig. 1). Both lobes comprise elongate, bifurcating, non‐segmented fin rays (lepidotrichia; Woodward, 1889a, 1889b, 1889c; Liston et al., 2013; Figure 1a); elements of the caudal skeleton supporting these fin‐rays in other bony fishes (epurals, hypurals, uroneurals) are not preserved in Leedsichthys (Liston et al., 2013), again, likely due to reduced skeletal ossification. In both teleost and non‐teleost ray‐finned fishes, the lepidotrichia are composed of a series of two opposing hemisegments of dermal bone (forming intramembranously), surrounding a space that contains blood vessels, fibroblasts (connective tissue cells), osteoblasts, osteoclasts, nerves and pigment cells (Akimenko et al., 2003; Arratia, 2008; Bercerra et al., 1983; Huang et al., 2003; Schultze & Arratia, 1989; Sousa et al., 2012; Zylberberg & Meunier, 2013). These hemisegments are held together by collagenous ligaments and connected by ligaments inserted into the lepidotrichium; the third set of ligaments runs between the lepidotrichia (Bercerra et al., 1983: fig. 6). By comparison, the caudal‐fin rays of Leedsichthys lack the hemisegments, with a histology including multiple concentric layers (lines of arrested growth [LAGs] or annuli) interpreted as representing the annual growth of the individual (Liston et al., 2013: fig. 10). These caudal‐fin rays are also unsegmented, with individual segments presumed to fuse together along the whole ray early in development (Figure 1). Lambers first identified the variability of segmentation in the caudal fins of pachycormids in his reappraisal of the group's phylogeny (Lambers, 1992). Of the pachycormids for which an assessment of the caudal fins can be made, 7 out of 14 genera have unsegmented caudal fin lobe lepidotrichia. This is a synapomorphy of the clade of edentulous pachycormids that includes Leedsichthys alongside other large, suspension‐feeding species (Arratia & Schultze, 2013; Dobson et al., 2019). Unpaired and unsegmented caudal‐fin ray‐like elements are also found as the ‘procurrent spines’ in some spiny‐rayed teleosts (Acanthomorpha), and spines are also characteristically found in the dorsal, anal, pectoral and pelvic fins in a range of other teleosts (e.g., Price et al., 2015).

FIGURE 1.

FIGURE 1

Leedsichthys problematicus Woodward 1889a (Peterborough, UK, Oxford clay, Middle Jurassic). (a–c) NHMUK PV P10000A. (a) Reconstructed caudal fin (2.74 m high), formerly exhibited at the Natural History Museum, London (note that identification of the upper and lower lobes is hypothetical [Liston & Noè, 2004]), (b, c) caudal fin lepidotrichia with rounded calluses (arrowheads); (b) multiple calluses in approximated position near the edge of the caudal fin; (c) closeup of calluses. Black arrow indicates anterior direction.

Additionally, in this specimen of Leedsichthys, rounded bony calluses are found on multiple caudal fin lepidotrichia (Figure 1b, c, white arrowheads; Liston, 2007; Liston et al., 2013; see also Liston, 2010 for a full listing of their occurrence across the hypodigm of Leedsichthys). These superficially resemble the calluses known as ‘Tilly bones’, also present on the postcranial skeleton of a range of teleost fishes and formed from an expansion (hyperostosis) of the external periosteal layer of the bone (Meunier et al., 2008; Paig‐Tran et al., 2016; Smith‐Vaniz et al., 1995; van Mesdag et al., 2020). The function of these Tilly bones is debated (see Paig‐Tran et al., 2016 for a recent discussion), but they are not thought to result from disease or damage (van Mesdag et al., 2020, fig. 1C), while the calluses associated with the Leedsichthys caudal fin lepidotrichia clearly surround an area of breakage (Liston, 2007; Liston et al., 2013). Moreover, hyperostotic bone is found in teleost bony elements of either the axial skeleton (vertebrae; Béarez et al., 2005; van Mesdag et al., 2020) or of the appendicular skeleton (pectoral fin, fin pterygiophores; Smith‐Vaniz et al., 1995; Meunier et al., 2010; Paig‐Tran et al., 2016; Chanet, 2018), and possibly of the cranial skeleton (van Mesdag et al., 2020, fig. 1F, G; Meunier & Béarez, 2021), but as far as we are aware not in the caudal‐fin rays themselves.

We describe in detail the histology of the osteocytic bone repair, in response to lepidotrichial damage in Leedsichthys. A callus develops around the region where the principal damage has occurred (e.g., breakage). Woven bone, with a high density of osteocytes, is deposited onto damaged bone surfaces, while circumvascular bone later forms around blood vessels as osteones. Resorption is evidenced by curved margins on the bone fragments prior to new bone forming, then, as resorption changes to deposition, resting lines in the bone microstructure show this change of activity.

2. MATERIALS AND METHODS

The caudal fin specimen (2.74 m high) from NHMUK PV P10000A was sent to the Natural History Museum, London (NHM), in March 1899 at the end of a year of reconstruction by Alfred Leeds, Mary Ferrier Leeds and Edward Thurlow Leeds, after some dispute with the museum concerning the packing required. The first confirmation of the caudal fin being on display is the 1905 museum guide, written by Arthur Smith Woodward. No images, beyond Alfred's correspondence sketches made during the period of the March 1898 excavation, appear to have existed prior to the photograph taken by the museum at Edward Thurlow Leeds' request in September 1937, 50 years before it was taken off display and put into storage. Comparison of this historical image with the specimen as recently reconstructed (Figure 1a) shows that some fin ray material has detached over the ensuing years, particularly in the posterior section of the inferior lobe, where the highest concentration of callus growths has been found (Liston et al., 2013).

Some of these bony calluses were returned to the specimen in 2018 (Figure 1b, c) and two were available for study: NHMUK PV P10000A(c), NHMUK PV P10000A(b). Both calluses were μCT scanned (Nikon Metrology HMX ST 225; voxel size 0.015, filter 1.5 mm copper, energy 130 kV and 115 μA, 3142 projections, exposure 708 ms), while NHMUK PV P10000A(b) was serial sectioned and a polished block produced (both at the Image and Analysis Centre, NHM). The slides and polished block were imaged using a Keyence VHX‐7000 digital microscope (Faculty of Dentistry, Oral & Craniofacial Sciences at King's College London, Guy's Campus; Figures 6, 7, 8), as well as an Olympus BX63 microscope with an Olympus DP73 camera and Cell Sens Dimension software (Sackler Imaging Suite, NHM; Figures 5, 8). Sections of undamaged lepidotrichia were also imaged (PCM F174; PETMG F174; NHMUK PV P6921) using the latter system. Both brightfield and polarized light were used. Lepidotrichia in Figures 2a and 3 are oriented so that the upper part in the image represents the proximal edge (closer to the articulation point within the caudal fin), while the lower part is more distal, closer to the fin margin. In Figures 4, 5, 6, 7, 8, the lepidotrichium has been rotated so that proximal is to the left and distal to the right. Figure 4 provides a map denoting the regions illustrated in Figures 5, 6, 7.

FIGURE 6.

FIGURE 6

Leedsichthys problematicus Woodward, 1889a (Peterborough, UK, Oxford clay, Middle Jurassic), NHMUK PV P10000A(b), (a–d) histology of damaged caudal fin lepidotrichium, brightfield light. See Figure 4 for the location of images on the lepidotrichium. (a) Series of openings for the vasculature, larger (newer) openings to the left of the image, (‘b’) above white line indicates region shown in (b); (b) five openings (foramina) for blood vessels, the foramina furthest to the left is largest (latest to develop), black asterisk indicates osteoid being deposited (also (c, d)). Woven bone also being deposited (middle foramen), with the two vascular canals furthest to the right surrounded by more regular collagen fibres, forming an osteon; (c) close‐up of developing osteon, including osteocytes with processes in the osteoid. A small resorptive surface is visible, with woven bone being deposited onto this surface; (d) multiple resorptive surfaces are visible, including one on the far right of the image, where collagen fibres are inserting, anchoring the developing osteon to the original lepidotrichium. Coll, collagen fibres; lepido, lepidotrichium; ost, osteon, ostc, osteocyte; res, resorptive surface; vc, openings for vasculature (blood vessels); wb, woven bone. Scale bars (a) = 1 mm, (b) = 100 μm; (c, d) = 20 μm.

FIGURE 7.

FIGURE 7

Leedsichthys problematicus Woodward, 1889a (Peterborough, UK, Oxford clay, Middle Jurassic), NHMUK PV P10000A(b), (a–d), histology of damaged caudal fin lepidotrichium, brightfield light. See Figure 4 for the location of images on the lepidotrichium. (a) large white arrowhead indicates major resorptive margin (also (c)), white boxes indicate areas in close‐up in (b, c); (b) close‐up of region of the callus very rich in vasculature, with blood vessels being surrounded by bone to form vascular canals. Small white arrowheads indicate resorptive margin, small black arrowheads mark regions where larger openings were being subdivided into smaller vascular canals, black asterisks indicate woven bone being deposited; (c) including large region of vasculature surrounded by woven bone, white arrows and arrowheads indicating resorptive surfaces; (d) close‐up of region indicated by white box in (c), showing differing orientation of woven bone fibres. br.b, broken bone; coll, collagen fibres; lepido, lepidotrichium; res, resorptive surface, vc, openings for vasculature (blood vessels); wb, woven bone. Scale bars (a) = 200 μm; (b, c, d) = 100 μm.

FIGURE 8.

FIGURE 8

Leedsichthys problematicus Woodward, 1889a (Peterborough, UK, Oxford clay, Middle Jurassic), NHMUK PV P10000A(b), (a–e) histology of damaged caudal fin lepidotrichium, brightfield light. (a) Polished block, white box indicates views in (b–e); (b–e) developing osteons, (c) white arrows indicate thick mineralized sheaths amidst layers of woven bone, (d, e) vascular canals and osteons amongst woven bone. br.b, broken bone; coll, collagen fibres; ost, osteon,; vc, openings for vasculature (blood vessels); wb, woven bone. Scale bars: (a) = 1.0 mm; (b, d, e) = 100 μm; (c) = 100 μm.

FIGURE 5.

FIGURE 5

Leedsichthys problematicus Woodward, 1889a (Peterborough, UK, Oxford clay, Middle Jurassic), NHMUK PV P10000A(b). (a–h) Histology of caudal fin lepidotrichium, brightfield light; (a–d) within the callus, (e, f) outside the main area of damage and callus (cf. Figure 4). Comparisons between regular bone at the lepidotrichial margins (e.g., black asterisks (a, c, e, f)), and more internal bone dominated by short fibre bundles (dark in these due to postmortem staining, white asterisks (a, c, e, f)). In (b, d) these fibre bundles can be associated with vascular canal spaces (numbers 2, 3, 4) or less so (number 1), and in (b), the haphazard arrangement of these fibres can be clearly seen. See Figure 4 for the location of images on the lepidotrichium. (a, e) box indicates area shown in (b, f). Scale bars (a, c) = 200 μm; (b, d) = 100 μm.

FIGURE 2.

FIGURE 2

Leedsichthys problematicus Woodward, 1889a (Peterborough, UK, Oxford clay, Middle Jurassic). (a) NHMUK PV P10000A(b), caudal fin lepidotrichium with callus, external view; (b–f) histology of undamaged caudal fin lepidotrichium. (b) PETMG F174, serial section showing cellular (osteocytic) bone layers; (c–f), NHMUK PV P6921; (c, d) transverse serial section showing histology of the undamaged lepidotrichium; (c) whole section, white boxes indicating close‐up of area shown in (d, e, f); (d) primary and secondary osteons, arrowheads indicating putative reversal margin. (b–d) Brightfield light; (e, f) close‐up showing wavy bony layers in (e) brightfield and (f) polarized light. In (e) arrows indicate spaces between layers created post‐mortem that are very bright in (f), indicating birefringence. In (f), the strong birefringence shows multiple fibre bundles passing through the bony layers, suggested to be Sharpey's fibres. Lepido, lepidotrichia; ostc, osteocyte; pr.Ost, primary osteon, s.ost, secondary osteon; vc, foramina for vascular canals. Scale bars equal (a, c) = 1.0 cm; (b) = 250 μm; (d) = 0.5 mm; (e–h) = 50 μm.

FIGURE 3.

FIGURE 3

Leedsichthys problematicus Woodward, 1889a (Peterborough, UK, Oxford clay, Middle Jurassic), caudal fin lepidotrichia with bone callus surrounding breakage, reconstructed CT scan and virtual sections from CT scan. (a–f) NHMUK PV P10000A(c); (a, b) callus and lepidotrichium in (presumed) dorsal and ventral views; (c, d) longitudinal virtual section showing break through the lepidotrichium and surrounding repair bone of the callus, white arrowhead in (d) shows resorption surface cutting across bone of the lepidotrichium; (e, f) transverse virtual sections showing repair bone and resorption surface (white arrowhead in (f)); (g–k) NHMUK PV P10000A(b); (g, h) callus and lepidotrichium in (presumed) dorsal and ventral views; (i–k) longitudinal virtual sections showing damage (crushing) to the lepidotrichium and surrounding repair bone of the callus. White arrowheads indicate concave surfaces in the bone, suggesting resorption, asterisk marks central open space within the lepidotrichium for passage of blood vessels and nerves (in i, j). bnd, boundary between callus and lepidotrichium; br, break of the lepidotrichium; cr, crushing of lepidotrichium; lepido, lepidotrichium; la.rep.bo, laminar repair bone. Scale bars equal 1.0 cm.

FIGURE 4.

FIGURE 4

Leedsichthys problematicus Woodward, 1889a (Peterborough, UK, Oxford Clay, Middle Jurassic). (a, b) NHMUK PV P10000A(b), histology of damaged caudal fin lepidotrichium, map to images in Figures 5, 6, 7, brightfield light. Black asterisks indicate broken bone within the main area of damage, white arrowhead indicates large resorptive surface shown in Figure 7(a, c); black arrowheads indicate areas in the callus which may be in the process of being resorbed. dam, region of damaged lepidotrichium; vc, openings for vasculature (blood vessels). Scale bar equals 1.0 cm.

Institutional Abbreviations: NHMUK, Natural History Museum, London, UK; PETMG, Peterborough Museum and Art Gallery, Peterborough, UK; MoR, Muséum de Rouen, Normandie, France.

3. RESULTS

3.1. Normal Leedsichthys caudal fin lepidotrichia ‐ histological and virtual sections

The normal (undamaged) histology of the Leedsichthys caudal fin lepidotrichia has been previously described (Bardet et al., 1993; Liston, 2010; Liston et al., 2013). As noted above, each lepidotrichium is an elongate, unsegmented structure (Liston et al., 2013, fig. 1; Figures 1, 2), internally comprising multiple layers of cellular bone (Figure 2b–f), characterized by numerous small osteocyte lacunae (Liston et al., 2013: fig. 9A, C; Davesne et al., 2019, 2020; Figure 2b). Although the layers are concentric, circumferential appositional bone, they are uneven and wave‐like in certain regions (Figure 2b, c, e). The lepidotrichia are made of compact bone, with few longitudinal vascular canals being present (Figure 2c, vc). At least some of these canals are surrounded by secondary bone deposition, as suggested by the putative reversal lines (Figure 2d, white arrowheads) that represent the boundary of this new bone that cross‐cuts centripetal primary bone layers (i.e., secondary osteons; Figure 2d, s. ost). This deposition of secondary bone around vascular canals results from the resorption of pre‐existing primary bone, which formed the primary osteons (Figure 2d, pr.ost). A small central space is partially visible in longitudinal section, originally accommodating the blood vessels and nerves that pass through the central space between the lepidotrichial hemisegments in other actinopterygians (Figure 3i,j, asterisk). In the transverse section of Figure 2c, this central cavity is entirely filled with compact bone. Under polarized light (Figure 2f, white arrows), straight, birefringent banded structures pass through the bony layers. These were thick, intrinsic collagen fibres, with some measuring 15–20 μm in width, relating to the ligaments running between the lepidotrichia within the fin (Bercerra et al., 1983). In contrast, a very thick, convoluted birefringence (Figure 2f) is at right angles to these extrinsic fibres and generally coincident with the arrested growth lines (Figure 2e, arrows). This is interpreted as exaggerated spaces due to post‐mortem sediment deposition in‐between surfaces of weaknesses resulting from pauses in regular growth during cyclical growth.

3.2. Damaged Leedsichthys caudal fin lepidotrichia and wound healing

Externally (from CT scan) and internally (via virtual sectioning) the damaged region of the lepidotrichium is indicated by a rounded irregular callus (Figures 2a, 3a,b,g,h). In both specimens (Figure 3), the callus is rounded on one side (Figure 3a, h), but on the opposing side, a large depression is present, which can also be seen in the virtual section (Figure 3b,g,i, dep). Here, differences in the damage to the two lepidotrichia are apparent. In NHMUK PV P10000A(c), sharp margins of broken bone are obvious (Figure 3c, brk) that are not seen in NHMUK PV P10000A(b) (Figure 3i–k); the latter is dominated by small broken pieces of bone. In NHMUK PV P10000A(c) distinct concavities, representing bays of erosion, are present along the broken bone surfaces (Figure 3d, f, white arrowheads); these cut across multiple layers of bone where resorption has occurred. There are also small resorption concavities associated with damaged bone in NHMUK PV P10000A(b) (Figure 3i,j, white arrowheads). An indeterminate boundary between the callus and lepidotrichium, referred to as an attachment junction, can also be seen in virtual section (Figure 3c,d,i,j, bnd).

A section taken from NHMUK PVP 10000A(b) shows the extent, location and histology of the damaged bone (Figure 4), including bone located outside the main area of damage (Figure 5e,f), new bone that has been deposited around the breakage to form the protective callus (Figures 5a–d, 6), and within the open central space representing the main damage to the lepidotrichium, with broken pieces of bone remaining within (Figures 7, 8). This lepidotrichium has a very distinct histology compared to undamaged specimens (Figure 2b–e). Within the callus (Figure 5a–d), fibre bundles dominate internally, compared to the more normal‐looking osteocytic cellular bone at the margins, which lacks these distinctively dark fibre bundles (Figure 5, black versus white asterisks, darker fibres stained by mineralization postmortem). Some of these bundles are associated with vascular canals (Figure 5b, d, numbers 2, 3), others less so (Figure 5b, number 1), including being associated with layers of new bone (Figure 5d, number 1). These fibres can be found as patches (Figure 5b, d, number 3) and in general show haphazard, overlapping orientations (Figure 5b, d, numbers 3, 4). The fibre bundles differ from those described above in the undamaged lepidotrichia (Figure 2e,f), in being shorter and much more prevalent throughout the damaged lepidotrichia. These fibre bundles are also present in bone that is more distant from the damaged region (Figures 4, 5), running through the bone in a disorganized manner, particularly in the middle of the lepidotrichium.

The main area of damage (Figure 4a, dam) is represented by a space in which small, irregularly shaped pieces of bone can be seen (black asterisks). The margins of this space are not as strongly concave as seen in Figure 3c but are still present (Figures 3i,j, 4a, white arrowhead). New bone is forming as part of the callus (Figures 4a, 5, 6, 7), within the damaged area (Figure 7a,b) and proximal to this area as well, within the lepidotrichium (e.g., Figure 7a,c,d).

Within the callus (Figures 4, vc, 6a), along the margin between the lepidotrichium and the callus, is a row of openings showing a gradation in size and shape, being smaller and rounder to the right of the image, but larger and more elongate to the left. These are interpreted as openings for vascular canals. Along the outer margins of each of these openings is a lighter‐coloured material (Figure 6b–d, black asterisks) containing cells with elongate processes (Figure 6c, ostc). Coarse‐fibered bone surrounds these openings (Figure 6b–d, wb), with fibre bundles clearly visible as thick, mineralized sheaths (the crystalline nature of the sheaths can be seen in Figure 8c, white arrows). Beneath some of these canal spaces, curved surfaces are visible, assumed to be due to bone resorption (Figure 6c,d, res), and, in some instances, thin fibres appear to be anchoring the developing bone to the resorbed surface (Figure 6d, see also Figure 7b, coll). By comparison, the bone surrounding the smaller, more regular openings is notably different, as the fibres are not held in large bundles (as seen in the woven bone) and the matrix is homogeneous around the canal (Figure 6c,d, coll). Cells with elongate processes are also visible here and arranged in a circular fashion around the canal space (Figure 6d, ostc).

Other parts of the callus (Figures 7a,b, 8) show similar series of canal openings, although not as linearly arranged as in Figure 6; in Figure 7a,c, the formation of these openings is more clearly shown to be the result of an elongate opening becoming separated into the smaller canal openings. The openings are also lined internally with the lighter‐coloured material (Figure 7b, black asterisks), which appears to be contributing to the separation of the vascular spaces (Figure 7b, black arrowheads). Some openings appear to be entirely surrounded by the coarse‐fibred bone (Figure 7b, wb), while in others, the bone comprising the unbundled fibres is present (Figures 7b, 8c–e, ost). New bone has also developed within the main area of damage, being deposited along a broken edge of the original lepidotrichial bone, forming a circular space; again small, scalloped resorption margins are present, to which the new bone attaches (Figure 7a,b, br.b, wb, vc, res).

The series of canal openings in the callus illustrated in Figure 7a,b is associated with a resorptive margin (Figure 7a,c, large white arrowheads) that cuts into a large, unusual region of bone forming within the lepidotrichium itself (Figure 7a,d). This large region consists of multiple layers of coarse‐fibered bone surrounding an elongate space that appears to be separating into smaller, individual spaces (Figure 7c). The orientation of the fibres differs markedly between the layers (Figure 7d). This new bone is also associated with a resorptive margin (Figure 7c, white arrows), and is itself being eroded by the extended resorption surface described above (Figures 4, 7, white arrowheads).

4. DISCUSSION

4.1. Repair process of dermal bones

Repair processes in endochondral bone, where bone ossification is performed within cartilage, have been reported extensively (Dittmer & Firth, 2017). Also, repair in dermal bones (bones with no cartilaginous precursor), like the skull roof and lepidotrichia, has been investigated in mammals such as the mouse (e.g., Wang et al., 2012, 2020), reptiles (Irwin & Ferguson, 1986), amphibians (Muñoz et al., 2018) and teleost fishes (Guertzen et al., 2014; Moss, 1962; Sousa et al., 2012; Witten & Huysseune, 2009). Repair includes several stages, namely inflammation and formation of a haematoma at the site of the injury, followed by woven bone formation. Woven bone is said to be weak (Ip et al., 2016) and composed of loosely arranged, disorganized collagen fibres (de Ricqlès et al., 1991; Francillon‐Vieillot et al., 1990; Prondvai et al., 2014). Because of this disorganization, under polarized light woven bone is distinguished from both parallel‐fibered and lamellar bone, with their more organized collagen and aligned hydroxyapatite crystals (Malluche & Faugere, 1986; Mitchell & van Heteren, 2015; Prondvai et al., 2014). During repair, woven bone is eventually removed via resorption and replaced by compact lamellar bone (Wang et al., 2020). For example, in the frog Xenopus tropicalis, Muñoz et al. (2018) found that repair bone deposited 15 days after the injury (hole drilled into the dermal bone of the skull) comprised woven bone with disorganized collagen fibres, comparable to rapidly deposited woven bone in mammalian endochondral bone repair. By 30 days post‐injury, the bone in Xenopus consisted of layers of parallel‐fibered collagen, similar to lamellar bone in mammals, more slowly deposited in repair.

Zebrafish (as a modern teleost more closely related to Leedsichthys) dermal fin rays or lepidotrichia show repair responses to both crushing (Sousa et al., 2012) and fracture (Guertzen et al., 2014). In these, the repair process develops from a multi‐layered epithelium over the damage site. A few days after fracture of the lepidotrichia, repair occurs at the site as a thickening of collagenous bone matrix, representing the first stage, later transformed into mature bone tissue (Guertzen et al., 2014). Non‐fractured bone remains alongside repair bone, and the broken lepidotrichia are never fully united (Guertzen et al., 2014). This mode of bone repair accords well with the calluses in Leedsichthys, on a macrostructural level, as both maintain a grossly enlarged area of repair bone, with fragmented bone remaining internally (e.g. Figures 1, 4, 7).

4.2. Leedsichthys lepidotrichia repair

In virtual sections, one lepidotrichium shows a clean break of the bone (NHMUK PV P10000A(c); Figure 3b), with sharp edges within the callus, and large concavities where resorption has occurred as part of the repair process (Figure 3d, white arrowheads). However, further deposition of repair bone is limited. In NHMUK PV P10000A(b) (Figure 3g–k), the damaged area includes more small pieces of a broken bone. The lack of cleanly broken surfaces here suggests that resorption has progressed further than in NHMUK PV P10000A(c), and multiple resorption surfaces have been identified, including an extensive surface described above, that is cutting into newly formed woven bone (Figures 4a, 7a,c, large white arrowhead).

Woven bone can be identified at several places in the damaged lepidotrichium, in conjunction with open spaces representing the presence of blood vessels entering the damaged region. The large number of vascular canal spaces within the callus indicates a rich extended blood supply with these being more (Figures 4a, 6) or less (Figure 7) organized. However, even when the vascular canal spaces are less well organized, they still appear to form a series, resulting from the division of a larger space into smaller units for individual blood vessels (e.g., Figures 7b, 8d). The woven bone that surrounds these canals can also be connected (anchored) by collagen fibre bundles running across the resting growth lines (Figures 6d, 7b, white arrowheads). Woven bone is recognizable by the mineralized linear microspaces in the fossil which held collagen fibre bundles in life (e.g., Figure 6c). These bundles are thick in the woven bone, and show a mixed orientation; often several layers of woven bone are present (Figures 7d, 8c).

The linear series of vascular canals in Figure 6 shows the development of the bone repair. Along the inner margin of the vascular canal, a lighter‐colour material is present that can be identified as osteoid, containing spaces representing large osteocyte lacunae, with elongate canaliculi (Figure 6c, ostc). Subsequently, a developing osteon is formed in the repair bone where the size of the vascular canal decreases and becomes rounder in section, with circumferentially‐organized collagen fibres present as fine‐fibered circumvascular bone with recognizable, large osteocytes (Figures 6d, ostc). This represents growing circumvascular bone. Thus, a developmental series can be proposed following damage: (1) an initial (hypothetical) inflammatory phase and formation of a haematoma (Maruyama et al., 2020, although not as large a response as in mammal repair [Moss, 1962; Herbst et al., 2019]); (2) blood vessels enter the area, with osteoclasts resorbing original bone to prepare a margin for new bone deposition (vascular response being crucial in mammalian bone healing; Tomlinson & Silva, 2013); (3) osteoblasts at this resting surface deposit woven bone that entraps the anchoring fibres, normal to these intrinsic fibres, and thicker; (4) woven bone surrounds vascular canals; (5) the size of the vascular canal decreases with layers of organised bone surrounding the canal space, representing the circumvascular bone; (6) all woven bone is eventually resorbed and only organised, fine fibered bone remains.

One of the more intriguing areas of repaired bone is shown in Figure 7a,c,d, a mixture of woven bone and primary osteons, the latter formed by deposition after resorption ceased, such that new growth lines intersect those of a large area of woven bone (Figure 7c,d). Here, multiple layers of coarse‐fibred woven bone are present, with differing fibre orientations, surrounding a relatively small elongate opening that appears to be subdividing into separate vascular canals. This may have been a large damaged area within the lepidotrichium that included multiple blood vessels, around which woven bone formed to stabilize the bone. Repair bone also occurs outside the main damaged area where short, thin, disorganized fibres are dominant, presumably to stabilize the bone here (e.g., Figure 5e,f). A comparable repair response has been previously described in ankylosaur osteoderms (Scheyer & Sander, 2003).

A similar repair process to Leedsichthys has been observed in a regeneration experiment on the caudal fin lepidotrichia of the teleost cichlid fish Tilapia (Kemp & Park, 1970), where mineralisation occurs after 7 days and continues as more and more of the collagen fibres (intrafibrillar) and the interfibrillar matrix show that the mineral as crystals of hydroxyapatite becomes extensive in all parts of the osteoid matrix. Here we observed the results of this process of dermal bone regeneration in a natural environment, where the damage of the Leedsichthys caudal fin has recovered to make a callus of repair bone joining all parts of the existing bone.

In Leedsichthys, although resorption occurs in the damaged area in conjunction with the broken lepidotrichium, resorption has not occurred to fully reduce the size of the callus. There are areas within the callus of NHMUK PV P10000A(b) that may have been resorbed (Figure 4a, black arrowheads), but a substantial portion of the callus remains. This would be comparable to the zebrafish lepidotrichium, also incompletely resorbed, leaving behind a swollen lepidotrichium (Sousa et al., 2012). These similarities are present, despite notable differences in the structure of the lepidotrichium between Leedsichthys and the zebrafish (lepidotrichia being paired structures in zebrafish, as in virtually all actinopterygians outside of edentulous pachycormids [Schultze & Arratia, 1989; Arratia, 2008; Liston et al., 2013]). Comparable bone repair occurs within tetrapod bone, suggesting conservation of the basic repair mechanism and healing processes in osteichthyans, at the cellular level. However, one important difference is the continued healing in tetrapod bone, such that early bone is completely replaced with lamellar bone, suggesting an increased strength of repaired bone (reviewed in Herbst et al., 2019).

Both specimens of Leedsichthys suggest ongoing repair in the lepidotrichia, lacking complete replacement of the damaged bone by compact lamellar bone; indeed, woven bone dominates, although osteons are beginning to develop (Figure 6c,d). In NHMUK PV P10000A(c) there is no repair to the damaged areas to knit them together, as the two broken margins of the bone remain apart, suggesting that it represents an early stage (e.g., haematoma), with less resorption overall in this callus, compared to NHMUK PV P10000A(b). The cause of the damage is more uncertain, as discussed below.

4.3. Compact, undamaged bone and osteons, relative to the physiology of Leedsichthys

In addition to the damage repair described above, one striking feature of Leedsichthys normal lepidotrichia is the occurrence of putative secondary osteons (Figure 2c,d). While bone resorption and remodelling are widespread in actinopterygians (Currey et al., 2017; Shahar & Dean, 2013; Witten & Huysseune, 2009), they are often less extensive than in tetrapods. Secondary osteons seem to be even rarer in ray‐finned fishes, although a complete survey of these, and other bone histological features, is lacking across the whole group.

In tetrapods, the occurrence of secondary osteons – and of extensive bone remodelling without damage to the bone – is associated with increased metabolic rates, most notably in mammals and birds (de Buffrénil & Quilhac, 2021). There is little to no comparative framing of actinopterygian bone histology relative to their physiology. Bone appears to be more intensely remodelled in large, fast‐swimming pelagic teleosts like tunas, the opah Lampris, and billfishes, taxa that are also capable of a form of endothermy through intense muscular activity and specialised metabolic pathways (Davesne et al., 2018; Legendre & Davesne, 2020). Secondary osteons are particularly prolific in one specialised structure, the rostrum of billfishes, which is filled with closely packed secondary osteons (Atkins et al., 2014). This can be related to microrepair of the elongate bill. Thus, it may also illuminate the process of repair in the broken lepidotrichium of Leedsichthys, remodelled by increased metabolism, as supported by high vascular density (Figure 6). Therefore, the proposed presence of secondary osteons in the repaired bone of Leedsichthys, although rare, may support the view that this animal had higher metabolic rates than most other actinopterygians, and was possibly endothermic like modern billfishes, tunas and opahs. This hypothesis has been already proposed – although not on the basis of bone histology – for Leedsichthys and the giant predator Xiphactinus from the Late Cretaceous (Ferrón, 2020; Ferrón et al., 2018). Further investigation of the histology of extant endothermic teleosts, and extinct taxa such as Leedsichthys and Xiphactinus, would justify the proposed relationship between bone structure and thermal physiology in actinopterygians.

4.4. Causes of lepidotrichial damage

One important ecological question we address is whether the repair response in the lepidotrichia of Leedsichthys described above was in response to breakage during swimming or damage due to predation. Although regarded as an open water suspension‐feeder, circumstantial evidence has also been used to argue that Leedsichthys also supplemented its diet durophagously, feeding on benthic prey (Liston, 2010) which might account for impacts on its environment. Both environmental and predatory causes are possible explanations given the peripheral position of the calluses on the trailing edge of the caudal fin, although the specific location of these is high in the mid‐section of the fin, rather than at the extreme apex of a lobe, makes breakage during swimming less likely. The sharp, straight edge of the damaged lepidotrichium in NHMUK PV P10000A(c) (Figure 3c) is plausibly due to breakage. In the other specimen (NHMUK PV P10000A(b), Figure 3i–k), breakage appears through the preserved part of the lepidotrichium, including outside the main damage (Figures 4, 5), although a clean break as in NHMUK PV P10000A(c) is absent. This damage instead may be due to predation, from crushing during capture, by several candidate taxa and indeed the localised clustering of these bony calluses suggests that they were derived from the same single cause, with only some of the fractured bones closer to the most intense part of the putative bite displaying crush, the more peripheral exhibiting a simpler fracture without crushing.

4.5. Predation on Leedsichthys

Of the roughly 70 known specimens of Leedsichthys, most appear to represent individuals that ranged from 8 to 12 metres in length (Liston et al., 2013), with only a small number of specimens showing possible traces of attack prior to our consideration of these lepidotrichia. There are certain advantages that Leedsichthys will have had over potential marine reptile predators once it achieved adult size. The depth of the body is estimated at 1.5–2.0 metres, given the height of the caudal fin (2.74 m for NHMUK PV P.10000A; Figure 1); such a deep body would have been an important factor in deterring gape‐limited predators (Domenici & Blake, 1997), constraining areas of the body that could be attacked. Of the two specimens that have been cited as evidence for predatory attack, both include a series of fin elements with marks that strongly resemble those of a bite. The first specimen (NHMUK PV P.6924; Liston, 2016) consists of a series of dorsal fin‐rays, and when the marks are reconstructed (Liston, 2007), provide an estimated bite width of 130 mm, with a tooth impression of 47 mm diameter. From these dimensions and the evidence for the specific mouth shape provided by the pattern of the bite, we agree with Liston (2007) that in the Oxford Clay seas this damage could only have been produced by the jaws and teeth of a pliosaur. Other potential predator taxa have jaws and teeth that are too dissimilar, and the plesiosauroids notably have too small a gape, with teeth well below the observed sizes of the dental impressions (Andrews, 1910).

The specimen NHMUK PV P.62054 also preserves a constrained arc of damage from an attacker in a collection of fin‐rays (probably pectoral). Again, these suggest the shape of a mouth, but smaller in size, possibly from a plesiosauroid or a small pliosaur (Andrews, 1913), being too broad and curved to have been made by a crocodilian.

Leedsichthys specimens discussed thus far indicate survival for an extended, but unknown, period of time after a predator attack. Skeletal damage could also be caused post‐mortem, via scavenging, where callus repair would be absent. For example, scavenging is the likely explanation for a Leedsichthys skull element (PETMG F.1), with an embedded Metriorhynchus sp. tooth (a thalattosuchian marine crocodylomorph). Although originally thought to result from predation, the purported bone growth around the tooth is minor and it appears that the bone was displaced by the entry of the tooth, rather than representing true bone repair (contra Martill, 1986). In addition, the bone is a hyomandibula, posterior to the jaws within the head region, a region less accessible to an attacking predator. Similarly, a Metriorhynchus superciliosus specimen (MoR 2012.4.67.80) from the Callovian of Normandie, France, contains a Leedsichthys gill raker within its stomach contents, showing further evidence that this group of marine reptiles scavenged on Leedsichthys. Other specimens of Leedsichthys exhibit indications of post‐mortem scavenging, such as hybodont (elasmobranch chondrichthyans) teeth embedded in the pectoral fins of NHMUK PV P.10000A (Liston et al., 2013: fig. 2B) and PETMG F.174 (pers. obs. JJL 2016).

Given the substantial response in the form of callus growth in the caudal fin lepidotrichia of NHMUK PV P.10000A, and in the dorsal fin rays of NHMUK PV P.6924, it is apparent that both are evidence of predatory attacks on Leedsichthys and that the intended prey survived, so scavenging can safely be discounted in these cases.

5. CONCLUSIONS

Two examples of dermal bone repair in caudal fin lepidotrichia of the stem‐teleost Leedsichthys show that the response is substantial and effective at joining the broken regions, by the formation of a callus. Repair also occurs away from this zone, where masses of unmineralized fibres stabilize the bone during life. In the repair zone, resorptive processes remove old bone, with new woven‐fibred bone rapidly deposited onto resorbed surfaces (allowing for rapid repair), changing the active process from osteoclastic resorption to deposition, represented by a reversal line. These reversal lines dominate the area under repair, where resorption prepares old bone for new coarse‐fibered bone, deposited around extrinsic attachment fibres, hence new bone is anchored to old bone via these peripheral fibres. Woven bone dominates, but with an organized vasculature. These vascular canals are lined with a thinner collagen fibre matrix representing circumvascular bone tissue, forming primary osteons. This vascular epithelium may contain the stem cells (neural crest derived, Cheah et al., 2010) that are osteogenic, essential to the repair response, making bone, both coarse‐ and fine‐fibered.

In Leedsichthys, the normal lepidotrichium is unpaired and mostly solid, lacking the hemisegments seen in other fishes. The space between these segments holds blood vessels, nevertheless, the preponderance of vascular canals in the damaged Leedsichthys lepidotrichium shows that a more than adequate supply was present. The large amount of woven bone present suggests that repair was still in its early phases in this individual, following an attack by a predatory marine reptile.

AUTHOR CONTRIBUTIONS

ZJ, JL and MMS conceived the study, MMS did the photography and ZJ segmented the CT‐scans (Figure 3), all authors contributed to the writing of the text.

ACKNOWLEDGEMENTS

We would like to thank Emma Bernard (Natural History Museum, London) for access to the fossil fish collection and for providing permission to serially section a Leedsichthys lepidotrichium. Callum Hatch and Tony Wighton are thanked for sectioning this lepidotrichium, and Vincent Fernandez and Brett Clark for CT scanning (all Core Research Labs, Natural History Museum, London). For microscopy, we thank Aaron LeBlanc, and Peter Pileki, FoDOCS, KCL for assistance with use of the Keyance Surface Illumination microscope. We also thank Stéphane Hua & Jérôme Tabouelle for bringing the Museum of Rouen specimen to our attention. Yara Haridy (Museum für Naturkunde, Berlin), Sophie Sanchez (Uppsala University) and an anonymous reviewer are thanked for reviewing an early version of the manuscript. In the Discussion, part of the section on predation is taken from chapter 9 of the unpublished thesis of JL.

Johanson, Z. , Liston, J. , Davesne, D. , Challands, T. & Meredith Smith, M. (2022) Mechanisms of dermal bone repair after predatory attack in the giant stem‐group teleost Leedsichthys problematicus Woodward, 1889a (Pachycormiformes). Journal of Anatomy, 241, 393–406. Available from: 10.1111/joa.13689

DATA AVAILABILITY STATEMENT

The data that support the findings of this study are available from the corresponding author upon reasonable request.

REFERENCES

  1. Akimenko, M.A. , Marí‐Beffa, M. , Becerra, J. & Géraudie, J. (2003) Old questions, new tools, and some answers to the mystery of fin regeneration. Developmental Dynamics, 226(2), 190–201. [DOI] [PubMed] [Google Scholar]
  2. Andrews, C.W. (1910) A descriptive catalogue of the marine reptiles of the Oxford clay, part 1. British Museum (Natural History), London, xvii + pp. 205. [Google Scholar]
  3. Andrews, C.W. (1913) A descriptive catalogue of the marine reptiles of the. Oxford Clay. Part II. London: British Museum (Natural History). [Google Scholar]
  4. Arratia, G. (2008) Actinopterygian postcranial skeleton with special reference to the diversity of fin ray elements, and the problem of identifying homologies. In: Arratia, G. , Schultze, H.‐P. & Wilson, M.V.H. (Eds.) Mesozoic fishes 4 – homology and phylogeny. München: Verlag Dr Friedrich Pfeil, pp. 49–101. [Google Scholar]
  5. Arratia, G. & Schultze, H.‐P. (2013) Outstanding features of a new late Jurassic pachycormiform fish from the Kimmeridgian of Brunn, Germany and comments on current understanding of pachycormiforms. In: Arratia, G. , Schultze, H.‐P. & Wilson, M.V.H. (Eds.) Mesozoic fishes 5 – global diversity and evolution. München: Verlag Dr Friedrich Pfeil, pp. 87–120. [Google Scholar]
  6. Atkins, A. , Dean, M.N. , Habegger, M.L. , Motta, P.J. , Ofer, L. , Repp, F. et al. (2014) Remodeling in bone without osteocytes: billfish challenge bone structure–function paradigms. Proceedings of the National Academy of Sciences, 111, 16047–16052. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Bardet, N. , Pennettier, E. , Pennetier, G. , Charles, A. & Charles, J. (1993) Des os énigmatiques à section triangulaire dans le Jurassique Moyen (Callovien) de Normandie. Bulletin trimestriel de la Société géologique de Normandie et des amis du Muséum du Havre, 80, 7–10. [Google Scholar]
  8. Béarez, P. , Meunier, F.J. & Kacem, A. (2005) Description morphologique et histologique de l'hyperostose vertébrale chez la thonine noire, Euthynnus lineatus (Teleostei: Perciformes: Scombridae). Cahiers de Biologie Marine, 46, 21–28. [Google Scholar]
  9. Bercerra, J. , Montes, G.S. , Bexiga, S.R.R. & Junqueira, L.C.U. (1983) Structure of the tail fin in teleosts. Cell and Tissue Research, 230, 127–137. [DOI] [PubMed] [Google Scholar]
  10. Chanet, B. (2018) Swollen bones in jacks and relatives (Teleostei: Acanthomorphata: Carangidae). Cybium, 42, 99–103. [Google Scholar]
  11. Cheah, F.S. , Winkler, C. , Jabs, E.W. & Chong, S.S. (2010) Tgfbeta3 regulation of chondrogenesis and osteogenesis in zebrafish is mediated through formation and survival of a subpopulation of the cranial neural crest. Mechanisms of Development, 127, 329–344. [DOI] [PubMed] [Google Scholar]
  12. Currey, J.D. , Dean, M.N. & Shahar, R. (2017) Revisiting the links between bone remodelling and osteocytes: insights from across phyla. Biological Reviews, 92, 1702–1719. [DOI] [PubMed] [Google Scholar]
  13. Davesne, D. , Meunier, F.J. , Friedman, M. , Benson, R.B.J. & Otero, O. (2018) Histology of the endothermic opah (Lampris sp.) suggests a new structure–function relationship in teleost fish bone. Biology Letters, 14, 20180270. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Davesne, D. , Meunier, F.J. , Schmitt, A.D. , Friedman, M. , Otero, O. & Benson, R.B.J. (2019) The phylogenetic origin and evolution of acellular bone in teleost fishes: insights into osteocyte function in bone metabolism. Biological Reviews, 94, 1338–1363. [DOI] [PubMed] [Google Scholar]
  15. Davesne, D. , Schmitt, A.D. , Fernandez, V. , Sanchez, S. & Benson, R.B.J. (2020) Three‐dimensional characterization of osteocyte volumes at multiple scales, and its relationship with bone biology and genome evolution in ray‐finned fishes. Journal of Evolutionary Biology, 33, 808–830. [DOI] [PubMed] [Google Scholar]
  16. de Buffrénil, V. & Quilhac, A. (2021) Bone remodelling. In: de Buffrénil, V. , Zylberberg, L. , de Ricqlès, A.J. , Padian, K. , Laurin, M. & Quilhac, A. (Eds.) Vertebrate skeletal histology and paleohistology. Boca Raton: CRC Press, pp. 229–246. [Google Scholar]
  17. de Ricqlès, A. , Meunier, F.J. , Castanet, J. & Francillon‐Vieillot, H. (1991) Comparative microstructure of bone. In: Hall, B.K. (Ed.) Bone: a treatise, Volume III. Boca Raton: CRC Press, pp. 1–78. [Google Scholar]
  18. Dittmer, K.E. & Firth, E.C. (2017) Mechanisms of bone response to injury. Journal of Veterinary Diagnostic Investigation, 29, 385–395. [DOI] [PubMed] [Google Scholar]
  19. Dobson, C. , Giles, S. , Johanson, Z. , Liston, J. & Friedman, M. (2019) Cranial osteology of the middle Jurassic (Callovian) Martillichthys renwickae (Neopterygii, Pachycormiformes) with comments on the evolution and ecology of edentulous pachycormiforms. Papers in Palaeontology, 2019, 1–26. [Google Scholar]
  20. Domenici, P. & Blake, R.W. (1997) Fish fast‐start kinematics and performance. Journal of Experimental Biology, 200, 1165–1178. [DOI] [PubMed] [Google Scholar]
  21. Ferrón, H.G. (2020) Evidence of endothermy in the extinct macropredatory osteichthyan Xiphactinus audax (Teleostei, Ichthyodectiformes). Journal of Vertebrate Paleontology, 39, e1724123. 10.1080/02724634.2019.1724123 [DOI] [Google Scholar]
  22. Ferrón, H.G. , Holgado, B. , Liston, J.J. , Martínez‐Pérez, C. & Botella, H. (2018) Assessing metabolic constraints on the maximum body size of actinopterygians: locomotion energetics of Leedsichthys problematicus (Actinopterygii: Pachycormiformes). Palaeontology, 61, 775–783. [Google Scholar]
  23. Francillon‐Vieillot, H. , de Buffrénil, V. , Castanet, J. , Géraudie, J. , Meunier, F.J. , Sire, J.‐Y. et al. (1990) Microstructure and mineralization of vertebrate skeletal tissues. In: Carter, J.G. (Ed.) Skeletal biomineralization: patterns, processes and evolutionary trends, Volume I. New York: Van Nostrand Reinhold, pp. 471–530. [Google Scholar]
  24. Friedman, M. , Shimada, K. , Everhart, M.J. , Irwin, K.J. , Grandstaff, B.S. & Stewart, J.D. (2013) Geographic and stratigraphic distribution of the Late Cretaceous suspension‐feeding bony fish Bonnerichthys gladius (Teleostei, Pachycormiformes). Journal of Vertebrate Paleontology, 33, 35–47. [Google Scholar]
  25. Friedman, M. , Shimada, K. , Martin, L.D. , Everhart, M.J. , Liston, J. , Maltese, A. et al. (2010) 100‐million‐year dynasty of giant planktivorous bony fishes in the Mesozoic seas. Science, 327, 990–993. [DOI] [PubMed] [Google Scholar]
  26. Guertzen, K. , Knopf, F. , Wehner, D. , Huitema, L.F.A. , Schulte‐Merker, S. & Weidinger, G. (2014) Mature osteoblasts dedifferentiate in response to traumatic bone injury in the zebrafish fin and skull. Development, 141, 2225–2234. [DOI] [PubMed] [Google Scholar]
  27. Herbst, E.C. , Doube, M. , Smithson, T.R. , Clack, J.A. & Hutchinson, J.R. (2019) Bony lesions in early tetrapods and the evolution of mineralized tissue repair. Paleobiology, 45, 676–697. [Google Scholar]
  28. Huang, C.‐C. , Lawson, N.D. , Weinstein, B.M. & Johnson, S.L. (2003) reg6 is required for branching morphogenesis during blood vessel regeneration in zebrafish caudal fins. Developmental Biology, 264, 263–274. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Ip, V. , Toth, Z. , Chibnall, J. & McBride‐Gagyi, S. (2016) Remnant woven bone and calcified cartilage in mouse bone: differences between ages/sex and effects on bone strength. PLoS One, 11(11), e0166476. 10.1371/journal.pone.0166476 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Irwin, C.R. & Ferguson, M.W. (1986) Fracture repair of reptilian dermal bone: can reptiles form secondary cartilage? Journal of Anatomy, 146, 53–64. [PMC free article] [PubMed] [Google Scholar]
  31. Kemp, N.E. & Park, J.H. (1970) Regeneration of lepidotrichia and actinotrichia in the tailfin of the teleost tilapia Mossambica . Developmental Biology, 22, 321–342. [DOI] [PubMed] [Google Scholar]
  32. Lambers, P. (1992) On the ichthyofauna of the Solnhofen lithographic limestone (upper Jurassic, Germany). Unpublished PhD, pp. 336. Rijksuniversiteit, Groningen. [Google Scholar]
  33. Legendre, L.J. & Davesne, D. (2020) The evolution of mechanisms involved in vertebrate endothermy. Philosophical Transactions of the Royal Society B: Biological Sciences, 375, 20190136. 10.1098/rstb.2019.0136 [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Liston, J.J. (2004) An overview of the pachycormiform Leedsichthys . In: Arratia, G. & Tintori, A. (Eds.) Mesozoic fishes 3‐ systematics, paleoenvironments and biodiversity. München: Verlag Dr. Friedrich Pfeil, pp. 379–390. [Google Scholar]
  35. Liston, J. J. (2007) A fish fit for Ozymandias? The ecology, growth and osteology of Leedsichthys (Pachycormidae, Actinopterygii). Unpublished PhD thesis, University of Glasgow. [Google Scholar]
  36. Liston, J.J. (2008) A review of the characters of the edentulous pachycormiforms Leedsichthys, Asthenocormus and Martillichthys nov. gen. In: Arratia, G. , Schultze, H.‐P. & Wilson, M.V.H. (Eds.) Mesozoic fishes 4 – homology and phylogeny. München: Verlag Dr Friedrich Pfeil, pp. 181–198. [Google Scholar]
  37. Liston, J.J. (2010) The occurrence of the middle Jurassic pachycormid fish Leedsichthys . Oryctos, 9, 1–36. [Google Scholar]
  38. Liston, J.J. (2016) Leedsichthys problematicus: Arthur Smith Woodward's 'most embarrassing enigma'. In: Johanson, Z. , Barrett, P.M. , Richter, M. & Smith, M. (Eds.) Arthur Smith Woodward: his life and influence on modern vertebrate Palaeontology. Special Publication, Vol. 430. London: Geological Society, pp. 235–259. [Google Scholar]
  39. Liston, J.J. , Newbrey, M.G. , Challands, T.J. & Adams, C.E. (2013) Growth, age and size of the Jurassic pachycormid Leedsichthys problematicus (Osteichthyes: Actinopterygii). In: Arratia, G. , Schultze, H.‐P. & Wilson, M.V.H. (Eds.) Mesozoic fishes 5 – global diversity and evolution. München: Verlag Pfeil, pp. 145–175. [Google Scholar]
  40. Liston, J.J. & Noè, L.F. (2004) The tail of the Jurassic fish Leedsichthys problematicus (Osteichthyes: Actinopterygii) collected by Alfred Nicholson Leeds ‐ an example of the importance of historical records in palaeontology. Archives of Natural History, 31, 236–252. [Google Scholar]
  41. Malluche, H.H. & Faugere, M.‐C. (1986) Atlas of mineralized bone histology. Basel: Karger, p. 136. [Google Scholar]
  42. Martill, D.M. (1986) The diet of Metriorhynchus, a Mesozoic marine crocodile. Neues Jahrbuch fur Geologie und Paläontologie, Monatshefte, 1986, 621–625. [Google Scholar]
  43. Maruyama, M. , Rhee, C. , Utsunomiya, T. , Zhang, N. , Ueno, M. , Yao, Z. et al. (2020) Modulation of the inflammatory response and bone healing. Frontiers in Endocrinology, 11, 386. 10.3389/fendo.2020.00386 [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Meunier, F.J. & Béarez, P. (2021) Bony tissues of the supra‐orbital in the milkfish Chanos chanos (Forsskål, 1775) (Chanidae: Gonorynchiformes). Cahiers de Biologie Marine, 62, 284–289. [Google Scholar]
  45. Meunier, F.J. , Deschamps, M. & Kacem, A. (2008) Le squelette des poissons téléostéens: structure, développement, physiologie, pathologie. Annales de la Société Zoologique de France, 133, 9–32. [Google Scholar]
  46. Meunier, F.J. , Gaudant, J. & Bonnelli, E. (2010) Morphological and histological study of the hyperostoses of Lepidopus albyi (Sauvage, 1870), a fossil Trichiuridae from the Tortonian (upper Miocene) of Piedmont (Italy). Cybium, 34, 293–301. [Google Scholar]
  47. Mitchell, J. & van Heteren, A.H. (2015) A literature review of the spatial organization of lamellar bone. Comptes Rendus Palevol, 15, 23–31. [Google Scholar]
  48. Moss, M.L. (1962) Studies of the acellular bone of teleost fish. II. Response to fracture under normal and acalcemic conditions. Acta Anatomica (Basel), 48, 46–60. [DOI] [PubMed] [Google Scholar]
  49. Muñoz, D. , Castillo, H. , Henriquez, J.P. & Marcellini, S. (2018) Bone regeneration after traumatic skull injury in Xenopus tropicalis . Mechanisms of Development, 154, 153–161. [DOI] [PubMed] [Google Scholar]
  50. Paig‐Tran, E.W.M. , Barrios, A.S. & Ferry, L.A. (2016) Presence of repeating hyperostotic bones in dorsal pterygiophores of the oarfish, Regalecus russellii . Journal of Anatomy, 229, 560–567. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Price, S.A. , Friedman, S.T. & Wainwright, P.C. (2015) How predation shaped fish: the impact of fin spines on body form evolution across teleosts. Proceedings of the Royal Society B, 282, 20151428. 10.1098/rspb.2015.1428 [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Prondvai, E. , Stein, K.H.W. , de Ricqlès, A. & Cubo, J. (2014) Development based revision of bone tissue classification: the importance of semantics for science. Biological Journal of the Linnean Society, 112, 799–816. [Google Scholar]
  53. Scheyer, T.M. & Sander, P.M. (2003) Histology of ankylosaur osteoderms: implications for systematics and function. Journal of Vertebrate Paleontology, 24, 874–893. [Google Scholar]
  54. Schultze, H.‐P. & Arratia, G. (1989) The composition of the caudal skeleton of teleosts (Actinopterygil: Osteichthyes). Zoological Journal of the Linnean Society, 97, 189–231. [Google Scholar]
  55. Schumacher, B.A. , Shimada, K. , Liston, J.J. & Maltese, A.E. (2016) Highly specialized suspension‐feeding bony fish Rhinconichthys (Actinopterygii: Pachycormiformes) from the mid‐cretaceous of the United States, England, and Japan. Cretaceous Research, 61, 71–85. [Google Scholar]
  56. Shahar, R. & Dean, M.N. (2013) The enigmas of bone without osteocytes. Bonekey Reports, 2, 343. [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Smith‐Vaniz, W.F. , Kaufman, L.S. & Glowacki, J. (1995) Species‐specific patterns of hyperostosis in marine teleost fishes. Marine Biology, 121, 573–580. [Google Scholar]
  58. Sousa, S. , Valero, F. & Jacinto, A. (2012) A new zebrafish bone crush injury model. Biology Open, 1, 915–921. [DOI] [PMC free article] [PubMed] [Google Scholar]
  59. Tomlinson, R.E. & Silva, M.J. (2013) Skeletal blood flow in bone repair and maintenance. Bone Research, 1, 311–322. [DOI] [PMC free article] [PubMed] [Google Scholar]
  60. van Mesdag, S.N.K. , den Blaauwen, J. , Dean, M.N. & Johanson, Z. (2020) Hyperossification in the vertebral column of Devonian placoderm fishes (Arthrodira). Journal of Vertebrate Paleontology, 41(1), e1766477. [Google Scholar]
  61. Wang, D. , Gilbert, J.R. , Cray, J.J., Jr. , Kubala, A.A. , Shaw, M.A. , Billiar, T.R. et al. (2012) Accelerated calvarial healing in mice lacking toll‐like receptor 4. PLoS One, 7(10), e46945. 10.1371/journal.pone.0046945 [DOI] [PMC free article] [PubMed] [Google Scholar]
  62. Wang, D. , Gilbert, J.R. , Zhang, X. , Zhao, B. , Ker, D.F.E. & Cooper, G.M. (2020) Calvarial versus long bone: implications for tailoring skeletal tissue engineering. Tissue Engineering: Part B, 26, 46–63. [DOI] [PubMed] [Google Scholar]
  63. Witten, P.E. & Huysseune, A. (2009) A comparative view on mechanisms and functions of skeletal remodelling in teleost fish, with special emphasis on osteoclasts and their function. Biological Reviews, 84, 315–346. [DOI] [PubMed] [Google Scholar]
  64. Woodward, A.S. (1889a) On the palaeontology of sturgeons. Proceedings of the Geological Association, 11, 24–44. [Google Scholar]
  65. Woodward, A.S. (1889b) Preliminary notes on some new and little‐known British Jurassic fishes. Geological Magazine Decade, 3(6), 448–455. [Google Scholar]
  66. Woodward, A.S. (1889c) Notes on some new and little‐known British Jurassic fishes. Annals and Magazine of Natural History Series, 6(4), 405–407. [Google Scholar]
  67. Zylberberg, L. & Meunier, F.J. (2013) Histology of Polypterus senegalus fin rays revisited. Comptes Rendus Palevol, 12, 203–210. [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.


Articles from Journal of Anatomy are provided here courtesy of Anatomical Society of Great Britain and Ireland

RESOURCES