Abstract

External and internal morphological characters of extant and fossil organisms are crucial to establishing their systematic position, ecological role and evolutionary trends. The lack of internal characters and soft-tissue preservation in many arthropod fossils, however, impedes comprehensive phylogenetic analyses and species descriptions according to taxonomic standards for Recent organisms. We found well-preserved three-dimensional anatomy in mineralized arthropods from Paleogene fissure fillings and demonstrate the value of these fossils by utilizing digitally reconstructed anatomical structure of a hister beetle. The new anatomical data facilitate a refinement of the species diagnosis and allowed us to reject a previous hypothesis of close phylogenetic relationship to an extant congeneric species. Our findings suggest that mineralized fossils, even those of macroscopically poor preservation, constitute a rich but yet largely unexploited source of anatomical data for fossil arthropods.

DOI: http://dx.doi.org/10.7554/eLife.12129.001

eLife digest

Fossils are the preserved remains of animals, plants or other organisms. The most highly prized fossils are those that retain their original three-dimensional shape and provide details needed to identify what species it represents, and what its closest living relatives might be. However, even fossils with the most beautifully preserved external anatomy can lack the internal structures that also help to identify its evolutionary history. “Mineralized” fossils are particularly useful for researchers as they form in a process that helps to preserve the internal anatomy of the organism. This type of fossil forms when mineral-laden water surrounds an organism’s body so that the minerals are deposited in its cells and turn soft tissues to stone.

In the 1940s, Swiss scientist Eduard Handschin used eight mineralized fossil specimens to describe a 25-40 million-year-old beetle species called Onthophilus intermedius. On the basis of the external anatomy of the two best-preserved specimens, Handschin claimed this species was distinct from, but closely related to a beetle species called O. striatus that is found in Europe today.

Since then, fossil examination methods have greatly advanced and include three-dimensional X-ray based imaging techniques that reveal the internal structures of a fossil while leaving it intact. One such technique is called X-ray computed tomography, in which numerous X-ray images of a solid object are taken from different angles. These images are then reassembled using computer software to create a virtual three-dimensional model of the object.

Here, Schwermann et al. used this X-ray technique to re-examine the beetle fossils originally reported by Handschin. This analysis revealed many new details of these specimens’ external and internal anatomies, including their gut, genitals and airways. These new insights place Onthophilus intermedius into a different evolutionary lineage to O. striatus. They also suggest that mineralized fossils could provide a rich source of data for studies on fossil insects and other arthropods, even if they appear to be poorly preserved on the outside.

DOI: http://dx.doi.org/10.7554/eLife.12129.002

Main text

Introduction

An organism’s morphology represents a complex solution to myriad ecological and environmental challenges it and its ancestors have confronted over evolutionary time. Inferring a comprehensive evolutionary history of a lineage requires consideration of a wide range of morphological features, and how they may have been shaped by selection, drift, and developmental constraints. While external characters predominate in ecomorphological and systematic studies, internal characters also play critical roles (Perreau and Tafforeau, 2011). In fossil specimens, however, these characters are usually not preserved or difficult to access (Siveter et al., 2007). While combined phylogenetic analyses of extant species frequently utilize internal anatomy, analyses including fossil taxa are generally limited to external characters. Moreover, it is often difficult to distinguish whether unobserved morphological characters were originally absent or lost due to taphonomic processes, potentially leading to misinterpretations of character evolution and erroneous phylogenetic placements (Sansom, 2015).

Several types of preservation or certain combinations of them are known for arthropod fossils. These are adpressions (compressions or impressions) (Wedmann et al., 2007; 2011), casts, voids, embeddings, mineral replications, charcoalified remains, or inclusions in amber (Grimaldi et al., 1994; Martı́nez-Delclòs et al., 2004; Grimaldi and Engel, 2005; Dunlop and Garwood 2014; Penney and Jepson, 2014). Amber inclusions are famous for exquisitely preserving three-dimensional external shape and sometimes internal characters (Perreau and Tafforeau, 2011). Three-dimensional arthropods may also be preserved within concretions (e.g. in siderite nodules [Nitecki, 1979; Garwood et al., 2009]), calcareous incrustations (e.g. in travertine [Rosendahl et al., 2013]), encapsulations in minerals (e.g. in onyx-marble [Pierce, 1951], chert [Anderson and Trewin, 2003], or gypsum crystals [Schlüter et al., 2003]), and mineral replications (e.g. as calcite [McCobb et al., 1998], silica [Miller and Lubkin, 2001], goethite [Grimaldi, 2009; Barling et al., 2014], pyrite [Grimaldi and Engel, 2005], or phosphate [Duncan and Briggs, 1996; Hellmund and Hellmund, 1996; Waloszek, 2003]). Some of these preservation types have revealed surprisingly detailed insights into the internal and soft tissue anatomy of several arthropods, for instance from several Paleozoic marine deposits (e.g. Siveter et al., 2007; 2013; 2014; Ma et al., 2014; Cong et al., 2014; Edgecombe et al., 2015). For insects, e.g. eyes (Duncan and Briggs, 1996) and muscle fibers (Grimaldi, 2009) have been reported.

Abundant arthropod fossils preserved by mineralization of calcium phosphate are known from the Oligocene fissure fillings of Ronheim (Hellmund and Hellmund, 1996), the Late Oligocene/Early Miocene limestones of Riversleigh (QLD, Australia) (Duncan and Briggs, 1996) and from Paleogene deposits at Quercy (south-central France) (Filhol, 1877; Gervais, 1877; Flach, 1890; Thévenin, 1903; Handschin, 1944). These localities have long been famous for their rich vertebrate fossils as well (e.g. Legendre et al., 1997; Laloy et al., 2013). The arthropod fossils of Quercy were documented by Swiss entomologist Eduard Handschin (1944). He described the hister beetle Onthophilus intermedius (Coleoptera: Histeridae) from eight specimens, and considered it distinct but closely related to the extant European species O. striatus (Forster, 1771). The description, however, was vague and based mainly on the external morphology of the two best-preserved specimens (Handschin, 1944).

X-ray microtomography has become established for the detailed examination of both extant (e.g. Betz et al., 2007; Bosselaers et al., 2010; van de Kamp et al., 2011; 2014; 2015; Brehm et al., 2015; Sombke et al., 2015) and extinct (Sutton, 2008; Sutton et al., 2014) arthropods, including fossils preserved in amber (Lak et al., 2009; Pohl et al., 2010; Soriano et al., 2010; Perreau and Tafforeau, 2011; Riedel et al., 2012). We explored the application of this technique to mineralized fossils by re-examination of Handschin's specimens of Onthophilus intermedius. To ensure a direct morphological comparison, we performed tomographic scans (Figure 1) of ethanol-fixed and air-dried O. striatus using the same experimental setup. Furthermore we tested the hypothesis that the two are closely related with a global phylogenetic analysis of Onthophilus Leach, 1817.

Figure 1.
Download figureOpen in new tabFigure 1. Comparison between the fossil Onthophilus intermedius (ADG) and EtOH-fixed (BEH) and air-dried (CF , I) specimens of O. striatus.

Slices of tomographic volumes showing head region (AC), thorax (D–F) and abdomen (G–I). ae = aedeagus; ag = accessory gland; bpae = basal part of aedeagus; hg = hindgut; m = musculature; ml = median lobe; mr = muscles remnants; mscx = mesocoxa; msf = mesofemur; mst = mesotibia; mt = muscle tissue; mtcx = metacoxa; mtf = metafemur; mtt = metatibia; pcx = procoxa; sph = spherical particle; sm = stony matrix; t8 = 8th abdominal tergite; t9 = 9th abdominal tergite; t10 = 10th abdominal tergite; te = tentorium; tr = trachea.

DOI: http://dx.doi.org/10.7554/eLife.12129.003

Results and discussion

We found internal characters in all fossils (Table 1). Three specimens show remains of inner organs, especially of the sclerotized genitalia, allowing their identification as two males and one female. The outer surfaces of most specimens appear smooth (Figure 2); the distinct punctuation found in extant Onthophilus species (Kovarik and Caterino, 2005) is faint.

Table 1.

Notes on the fossil Onthophilus intermedius specimens from Quercy and their preservation.

DOI: http://dx.doi.org/10.7554/eLife.12129.004

IDInternal structures preservedNotes
F1951some sclerites (incl. coxa-trochanteral joints) and tracheaethe only specimen depicted by Handschin (1944); but not explicitly designated as holotype
F1992some sclerites and small tracheaehead, prothorax missing
F1993some sclerites (incl. coxa-trochanteral joints)head, pygidia missing; elytra partly abraded
F1994most sclerites, muscle parts, tracheae, parts of alimentary system, large parts of male genitalsthe only specimen of the collection that is ventrally encrusted by a stone matrix
F1995some sclerites, parts of male genitalshead present; abdomen deeply abraded dorsally
F1996some scleriteshead, prothorax missing
F1997some sclerites, remains of muscles below the elytrahead, prothorax partly abraded
F1998some sclerites (incl. coxa-trochanteral joints), parts of female genitaliahead, prothorax partly abraded
Figure 2.
Download figureOpen in new tabFigure 2. Surface renderings of the eight Onthophilus intermedius specimens.

Note the unique encrustation of F1994.

DOI: http://dx.doi.org/10.7554/eLife.12129.005

The specimen F1994 (Figures 1A,D,G, 2, 3, Supplementary file 1) differs from all other samples by the presence of a stony matrix, covering the ventral part of the beetle. Its dorsal part and head are exposed; the elytra are missing and were probably detached before embedding. The exposed surface is partly eroded, especially in the anterior region of the head, and no appendages are visible from the outside. The matrix, however, concealed the best-preserved fossil from the collection, which we examine here in detail.

Figure 3.
Download figureOpen in new tabFigure 3. Digital reconstruction of the fossil.

(A) Photograph of Onthophilus intermedius (F1994) ventrally embedded in a stony matrix. (B) Digital reconstruction showing fossilized beetle (green) and matrix (brown). (C) Beetle digitally isolated from the stone, revealing well-preserved morphology hidden by the matrix. (D) Perspective view of the fossil showing parts of exoskeleton, tracheal network, alimentary canal and genitals. (E, F) Comparison of the male genitals of the extant O. striatus (E) and the fossil O. intermedius (F); outer sclerites cut to reveal internal anatomy. See Supplementary file 1 for an interactive version of the 3D reconstruction.

DOI: http://dx.doi.org/10.7554/eLife.12129.006

The ventral portion of the beetle covered by the matrix reveals an extraordinary preservation of exoskeletal fine structures and internal anatomy (Figures 3, 4 and 5; Supplementary file 1). While some fractions of the cuticle appear to be mineralized, the latter is mostly represented by air-filled spaces in the fossil (Figure 1A,D,G). The surface of the exoskeleton is preserved as a three-dimensional imprint of remarkable detail; the body sclerites show characteristic punctuation of the genus. The right foreleg is not preserved; the left one is truncated from the trochanter; distal parts of the leg were lost prior to fossilization. The right mid and hind legs are eroded at the edge of the matrix, but their encrusted left counterparts appear complete except for the most distal part of the metafemur of the hind leg that would protrude from the matrix. Moreover, many anatomical characters can be recognized inside the fossil (Figure 3D). Apart from internal invaginations of the exoskeleton (e.g. tentorium, furcal arms and metendosternite), large parts of the alimentary canal and tracheal system are visible. The oesophagus appears to be shrunken and is connected to the crop, which is truncated posteriorly. The anterior part of the hindgut is hollow, while the middle part is apparently filled with mineral matrix but well-defined. Conspicuous spherical particles may constitute remnants of gut content (Figure 1G). The hindmost part of the gut can be roughly retraced by aggregations of tiny holes inside the mineral matrix. Like in the alimentary canal, some large tracheae appear to be filled with matrix, while others are hollow. Except for the musculature connecting the right pro- and mesofurcal arms (Figure 1D), most muscles can only be recognized by remnants at the insertion areas (Figure 1G). The genitals are extraordinary well-preserved (Figure 3F). While testes and Ductus ejaculatorius could not be recognized, other soft tissues such as the spiral accessory glands and parts of the gland ducts are conspicuous. The genital sclerites, including aedeagus, median lobe, gonopore, tergites 8-10 and sternites 8 & 9 are almost perfectly preserved as imprints.

Figure 4.
Download figureOpen in new tabFigure 4. Coxa-trochanteral joints.

Comparison of the joints (cut) of the left mid- (AB) and hind leg (CD) of Onthophilus striatus (AC) and O. intermedius (BD), showing coxae (green) and trochanters (yellow).

DOI: http://dx.doi.org/10.7554/eLife.12129.007

Figure 5.
Download figureOpen in new tabFigure 5. Digital endocast of Onthophilus intermedius (specimen F1994).

A digital endocast (AB) artificially created from tomography data resembles the shape of the other fossils (Figure 2) much closer than the original surface of the beetle (CD) hidden by the stony matrix.

DOI: http://dx.doi.org/10.7554/eLife.12129.008

The remarkable preservation state of the fossil is emphasized when its morphological characters are compared to those of an extant ethanol-fixed specimen of the same genus (Figures 1, 3E,F and 4). The new anatomical data from this specimen facilitated an extended description of the species according to modern taxonomic standards (Appendix 1).

Handschin (1944) hypothesized a close relationship (‘particularly striking similarity’) between Onthophilus intermedius and O. striatus based on then-observable external morphology. However, phylogenetic analysis (Material and methods) of the more diverse character set now accessible places these species in distinct clades. The analysis resulted in 72 most parsimonious trees of length 185 (CI 0.27, RI 0.61). The strict consensus of these trees (Figure 6) is well resolved apart from a few rearrangements of some outgroup taxa and within a relatively derived group related to O. niponensis Lewis, 1907. O. intermedius is part of a trichotomy involving O. silvae Lewis, 1884 and a large group of species descended from the common ancestor of O. giganteus Heleva, 1978 and O. niponensis, though in reweighted trees it is resolved as sister to O. silvae alone. In all analyses O. striatus is nested within a lineage of Nearctic and far-eastern Palaearctic species, including O. flavicornis Lewis, 1884, O. flohri Lewis, 1888 and others.

Figure 6.
Download figureOpen in new tabFigure 6. Strict consensus tree.

The analysis places Onthophilus striatus within a lineage of Nearctic and far-eastern Palaearctic species (red), while O. intermedius is a member of a separate Holarctic lineage (blue). Four internal (purple) and three external (orange) unambiguous synapomorphies supporting their respective placements are mapped onto the cladogram - Onthophilus striatus group: Character 22:2, mesoventrite wide and short; 30:1, pygidial median carina absent; 35:2, tegmen of aedeagus abruptly downturned apically. O. intermedius group: 29:2, pygidium laterally impunctate; 36:2, tegmen of aedeagus abruptly narrowing apically; 40:2, lateral halves of eighth sternite large and nearly meeting at midline; 41:2, stem of spiculum gastrale broad throughout its length.

DOI: http://dx.doi.org/10.7554/eLife.12129.009

Inclusion of diverse characters revealed by microtomography of Onthophilus intermedius yields a well-supported topology and a more comprehensive picture of the biogeographic and morphological history of the group. Of the characters scored for both O. intermedius and O. striatus, there are seven by which their states differ, three external and four internal. Of these, two external (chars. 29 & 30) and one internal (char. 36) are reconstructed as autapomorphies (Figure 6). Only one external synapomorphy (char. 22) separates them, while three of the four genitalic differences (chars. 35, 40, and 41) represent synapomorphies of their respective lineages. Exclusion of internal characters for O. intermedius did not affect the topology, but did prevent genitalic characters from supporting its larger containing clade. Critical diagnostic differences in external morphology, such as mesoventral proportions and pygidial sculpturing, were also revealed by visualization of features previously obscured by matrix.

Based on our examinations we can reconstruct the probable fossilization process of the Quercy Onthophilus specimens, which culminates in a partial mineralization of inner organs in combination with the cuticle preserved as voids. An accurate three-dimensional conservation of soft tissues does not occur if the specimens are dried in air (Figure 1C,F,I). Therefore, the fixation process must have occurred fast, possibly due to the animal being immediately penetrated and enclosed by phosphate rich water. In arthropods, this type of fossilization is only known from a handful of localities, which are better known for a rich vertebrate fauna (Riversleigh: Duncan and Briggs, 1996; Ronheim: Hellmund and Hellmund, 1996; Quercy: Handschin, 1944). Replication of soft tissues by phosphatization may be accomplished over a period of weeks (Martı́nez-Delclòs et al., 2004). Possible sources for high phosphorous concentrations in water circulating through the fissure fill are rocks or abundant phosphate-rich vertebrate bones, which may have been deposited along with them (Handschin, 1944; Hellmund and Hellmund, 1996). After encrustation and internal mineralization, the cuticle largely decayed, leaving air-filled spaces. Erosion processes probably removed the outer stony matrix of most specimens, including fragile appendages and the imprint of the outer surface of the exoskeleton, leaving a mineralized endocast. Thus, the exterior of the fossils merely represents the inner surface of the exoskeleton – the deep grooves (Figure 2) actually being inner folds or apophyses. While the smooth dorsal part of F1994 resembles the other fossils in appearance, its ventral surface covered by the mineral matrix shows a distinct surface sculpturing as present in extant species of the genus. In contrast, an artificial ‘digital endocast’ created from the tomographic data of F1994 (Material and methods) bears a striking resemblance to the other fossils (Figure 5), on which Handschin based his original description. Summing up, the Quercy hister beetles represent three-dimensional ‘hybrid’ fossils, comprising cuticle imprints and mineralized soft tissue, combining to preserve both exoskeletal fine structure and internal anatomical characters.

Fissure filling fossils preserving three-dimensional internal anatomy will help to overcome taphonomic biases in available fossil data (Allison and Bottjer, 2011). To date, fossilized insect internal character information has mainly been obtained from well-preserved amber inclusions (e.g. Pohl, et al., 2010; Perreau and Tafforeau, 2011). However, the origination of amber as tree resin causes a representational bias toward generally arboreal taxa (Martı́nez-Delclòs et al., 2004). The fossil arthropods of Quercy represent an assemblage of taxa more typically associated with forest floor communities (Handschin, 1944), as exemplified by Onthophilus, typically a predator in various decaying organic materials (Kovarik and Caterino, 2005; Bajerlein et al., 2011). Such communities are less commonly preserved than those of many other environments (Kidwell and Flessa, 1996). Beyond anatomical data on these species, clearer interpretations of evolutionary relationships of these fossils will improve inferences about the evolution of these ecological communities. Thus, reexamination of the Quercy fossils, and likely also of similar mineralized fossils from other localities (which may represent different ecosystems and/or time periods), may provide a highly complementary source of information on the evolutionary history of arthropods.

With regard to the methods employed here, we can offer some guidance on improving future imaging attempts on similar materials. Based on our experience, a fast tomography setup combining filtered polychromatic radiation and an optimized detector system (dos Santos Rolo et al., 2014) is well-suited to achieve sufficient image quality in most fossil specimens. Thus, scan duration per tomogram may be reduced to a couple of seconds (Material and methods), facilitating high-throughput screening of large sample numbers in short time.

Our results demonstrate that mineralized arthropod fossils from a sedimentary context may three-dimensionally preserve soft tissue and other internal anatomical characters in remarkable detail, which allows determinations and phylogenetic analyses according to the standards for Recent organisms. Reevaluation of relationships with modern taxa in this extended morphological context will substantially improve estimates of rates and modes of arthropod evolution. This exceptionally detailed preservation may be aided by the presence of a surrounding stony matrix, hinting that encrusted specimens, which therefore were originally considered to be of poor quality, could contain particularly well-preserved external and internal characters. Our findings may trigger the reinvestigation of numerous similar fossils from various localities.

Materials and methods

Synchrotron X-ray microtomography

3D X-ray micro-computed tomography scans with synchrotron radiation (µCT) were performed at the TOPO-TOMO beamline (Rack et al., 2009) of the ANKA Synchrotron Radiation Facility at Karlsruhe Institute of Technology (KIT). The measurements consisted of the acquisition of 2500 equiangularly spaced radiographic projections of the sample in a range of 180°. The frame rate was set to 150 images per second, resulting in an overall scan duration of 16.67 seconds per sample. The parallel polychromatic X-ray beam produced by a 1.5 T bending magnet was spectrally filtered by 0.2 mm aluminum to obtain a peak at about 15 keV. The sample was placed 20 cm upstream of the detector, which in turn was located about 33 m from the source. The detector consists of a thin, plan-parallel lutetium aluminum garnet single crystal scintillator doped with cerium (LuAG:Ce), optically coupled via a Nikon Nikkor 85/1.4 photo-lens to a pco.dimax camera with a pixel matrix of 2008x2008 pixels. The lens was stopped down to F/4 to remove optical aberrations and to increase its depth of focus, permitting the use of a thicker scintillator to collect a higher fraction of the incident X-ray photons. The magnification of the optical system was adjusted to 3X, yielding an effective X-ray pixel size of 3.66 µm (dos Santos Rolo et al., 2014). Tomographic reconstruction was performed with the GPU-accelerated filtered back projection algorithm implemented in the software framework UFO (Vogelgesang et al., 2012). Microtomographic image data are deposited in Morph·D·Base (www.morphdbase.de; accession numbers T_vandeKamp_20151216-M-12.1 to T_vandeKamp_20151216-M-22.1).

3D reconstructions

3D reconstruction followed the protocol described by Ruthensteiner and Heß (2008) and van de Kamp et al., (2014); using Amira (versions 5.5, 6, FEI) and Avizo (version 8.1, FEI) for segmentation of the tomographic volumes and CINEMA 4D R15 (Maxon Computer GmbH) for assembly of components and rendering of figures. The ‘digital endocast’ (Figure 5) was created from the tomographic stack of specimen F1994 by segmenting solely the dorsal stony matrix, ventrally confined by the inner impression of the beetle’s cuticle.

The number of surface polygons was reduced to 10% of its original value in CINEMA 4D: the raw mesh of F1994 contains approx. 30 million polygons, the reduced version (Figure 3D) ca. 3 million. Segmentation artifacts were carefully removed using the sculpting tools of the software. For the interactive 3D model (Supplementary file 1), the polygon count was further reduced to 800,000 (without the stony matrix); the digital mesh was imported into Deep Exploration (version 6; Right Hemisphere), saved as Universal 3D file (U3D) and embedded into a PDF document with Adobe® Acrobat® 9 Pro Extended.

Phylogenetic analysis of Onthophilus intermedius

Our phylogenetic analysis was performed to test Handschin’s (1944) hypothesis of a close relationship of Onthophilus intermedius to the extant and sympatric O. striatus. Although his hypothesis was not presented in strictly phylogenetic terms (‘particularly striking similarity’; our translation), the suggestion is of a direct lineal relationship between these heterochronic species. This would be revealed in a cladistic analysis as a sister group relationship between them. Thus, the hypothesis would be rejected by any resolution in which O. intermedius and O. striatus were not found to be sister species. We compiled a character set comprising 41 characters (Source code 1) of internal and external morphology visible in one or more specimens of O. intermedius, as visualized following X-ray microtomography. We scored these characters for a set of 29 of the 39 currently described species in the genus Onthophilus (Mazur, 2011), as well as seven outgroup Onthophilinae (including the recently described Cretaceous Cretonthophilus tuberculatus (Caterino et al., 2015). Most were scored from direct examination of specimens. However, some taxa were scored from illustrations and descriptions in the literature (Reichardt, 1941; Helava and Howden, 1977; Helava, 1978; Ôhara and Nakane, 1986; Ôhara, 1989; Howden and Laplante, 2003).

Characters and states

  1. Sutures separating antennomeres of antennal club: 1, distinct; 2, indistinct.

  2. Position of antennal insertion: 1, at upper edge of eye; 2, in front of middle of eye.

  3. Proximity of antennal fovea and eye: 1, antennal fovea in contact with inner edge of eye; 2, separated by cuticular ridge from eye.

  4. Median frontal carina: 1, absent; 2, present.

  5. V-shaped lateral frontal carinae: 1, absent; 2, present.

  6. Labral setae: 1, bisetose; 2, plurisetose (due to secondary setae).

  7. Number of pronotal carinae: 1, zero; 2, two; 3. four; 4, six.

  8. Form of outer pronotal carina: 1, absent; 2, excavate along inner edge; 3, raised to form a simple carina.

  9. Completeness of outer pronotal carinae: 1, complete; 2, anteriorly abbreviated; 3, interrupted; 4, absent.

  10. Completeness of median pronotal carinae: 1, complete; 2, abbreviated; 3, interrupted; 4, absent.

  11. Consistency of strength of pronotal carinae: 1. all pronotal carinae equal strength; 2, pronotal carinae alternating in strength

  12. Pronotal sculpturing: 1, ground punctation absent; 2, simply punctate (finely or deeply); 3, surface reticulo-strigose (punctures elongated and dense).

  13. Lateral margin of pronotum: 1, without dense border of punctures along margin; 2, deeply punctate along inner edge of lateral margin.

  14. Longitudinal elytral carinae: 1, absent; 2, present.

  15. Evenness of elytral carinae: 1, elytral carinae similar in height; 2, alternating in height.

  16. Completeness of elytral carinae: 1, All complete; 2, One or more carinae interrupted along its length.

  17. Basal elytral foveae (between costae 2 & 4, sensu Helava (1978)): 1, without deep basal foveae; 2, with deep basal foveae.

  18. Foveae of elytral interstriae: 1, absent; 2, weak; 3, strong.

  19. Basal emargination of prosternal keel: 1, not emarginate, truncate or projecting; 2, narrowly, subacutely emarginate; 3, broadly, more obtusely emarginate.

  20. Lateral notch of prosternal lobe: 1, without lateral notch; 2, with lateral notch.

  21. Spination of protibia: 1, not densely spinose; 2, densely spinose.

  22. Proportions of mesoventrite: 1, nearly half as long as wide (ie. length/width ratio ~0.5); 2, wide and short (length/width ratio >0.5).

  23. Postmesocoxal stria of metaventrite: 1, absent (or totally obscured by punctures); 2, present.

  24. Punctation of metaventral disk: 1, uniform; 2, with discrete impunctate areas on either side of midline.

  25. Spination of outer margin of mesotibia: 1, absent; 2, present.

  26. Apical lateral spine of mesotibia: 1, absent or weakly produced, not disrupting outer margin of tibia; 2, Well developed, tibial apex produced.

  27. Median carina of propygidium: 1, absent; 2, present.

  28. Lateral carinae of propygidium: 1, absent; 2, present.

  29. Punctation of pygidium: 1, uniform; 2, with discrete impunctate areas on either side of midline.

  30. Median longitudinal carina of pygidium: 1, absent; 2, present, single; 3, present, doubled.

  31. Transverse carina of pygidium: 1, absent; 2, present.

  32. Basal piece, closure: 1, open, not forming a closed cylinder; 2, forming a complete, closed cylinder; 3, fused with tegmen (some Epiechinus only).

  33. Basal piece, length relative to tegmen: 1, long, nearly half length of tegmen; 2, much less than half length of tegmen.

  34. Tegmen midline division: 1, divided along entire midline to base; 2, fused along >1/4 of its length.

  35. Tegmen, apical curvature: 1, evenly curved to tip; 2, abruptly downturned at apex.

  36. Tegmen, height (as seen in lateral aspect): 1, evenly narrowing; 2, abruptly narrowing near midpoint.

  37. Tegmen, relative widths along length: 1, widest in basal half; 2, parallel-sided or widest in apical half.

  38. Point of median lobe extrusion (following Helava (1978)): 1, near dorsal apex; 2, ventrally, subapical.

  39. Tegmen, apices: 1, apices convergent; 2, apices parallel (approximate or separate); 3, apices divergent

  40. Development of 8th sternite: 1. lateral halves reduced, broadly separated; 2. halves more substantial, approaching or meeting at midline.

  41. Stem of 9th sternite (spiculum gastrale): 1, stem narrow, abruptly widened to apex; 2, stem broad, weakly widened to apex.

Data were analyzed under parsimony using PAUP* 4.0a144 (Swofford, 2002), using a heuristic search with 1000 random addition sequence replicates. Characters were all treated as unordered. We examined the effects of character reweighting (by rescaled consistency indices), and exclusion of various character subsets (internal vs. external). Character transitions were mapped using Mesquite v. 3.03 (Maddison and Maddison, 2015). The tree was rooted with either Anapleus (Dendrophilinae: Anapleini), considered to exhibit plesiomorphic states in many higher level histerid characters (Caterino and Vogler, 2002), or Cretonthophilus, a recently described taxon from Cretaceous Burmese amber representing the oldest known Onthophiline histerid (Caterino et al., 2015) (Source code 1).

Appendix 1

Redescription of Onthophilus intermedius Handschin, 1944

Figures 25, Appendix figure 1 and Supplementary file 1

Appendix figure 1.
Download figureOpen in new tabAppendix figure 1. Morphological characters visible in specimens other than F1994.

(A) Dorsal view of F1951, showing elytral carinae and foveae. (B) Posterolateral view of F1997, showing propygidial and pygidial carinae. (C) Posterior view of F1951. (D) Anterior view of F1995. (E) Valvifer and coxite of female ovipositor.

DOI: http://dx.doi.org/10.7554/eLife.12129.012

Type locality

"Larnagol (Quercy) 1902. Coll. Rossignol" (Handschin, 1944). Fissure fillings from Larnagol are not known. It seems, then, that this is a general reference to the fossils that were collected by Rossignol, residing in Larnagol.

Type material

Lectotype male, here designated (housed in Natural History Museum of Basel): Specimen F1994, though largely encased in stone matrix, uniquely preserves external and internal morphology suitable for species diagnosis. Handschin (1944) explicitly based his description on the two best preserved specimens out of eight, without identifying them or selecting any one as a primary type. We here specify a lectotype due to the highly variable state of preservation of the material available, and considerable possibility of misinterpretation of what are mostly endocasts.

Other material

Paralectotypes: male (F1995); female (F1998); undetermined sex (F1951, F1992, F1993, F1996, F1997).

Description

Length: 4.6 mm, width: 2.8 mm; body elongate oval, distinctly costate on dorsal surfaces; head with convergent frontal carinae (Appendix figure 1). Pronotum with six uninterrupted costae; outer and lateral costae (PC1 & PC2 sensu Helava (1978) slightly abbreviated, obsolete in anterior one-third; median costae (PC3) complete (Appendix figure 1D); lateral pronotal margin slightly elevated, strongly arcuately narrowed from base to apex. Elytra each with three strong, complete dorsal costae (ISC, EC2, & EC4 sensu Helava (1978)), EC1, EC3, and EC5 only weakly developed (Appendix figure 1A); striae deeply punctured along their lengths. Propygidium about twice as wide as midline length, depressed along anterior margin; disk with distinct median and lateral carinae (PMC and PLC sensu Helava (1978)), median carina most strongly produced just behind middle, rapidly diminshing anteriorly and posteriorly; lateral carinae weaker and short, little more than lateral tubercles (Appendix figure 1B,C); Pygidium slightly longer than wide, with distinct longitudinal and transverse carinae (LC and TC sensu Helava (1978)), the longitudinal carina varied in strength, appearing more complete in endocasts; transverse carina complete, slightly expanded at lateral extremes; pygidial disk conspicuously punctate, with punctures slightly smaller distad, with two small impunctate areas on either side of midline. Prosternal keel emarginate at base (Appendix figure 1D); prosternal lobe short; antennal cavities present in anterior corners of hypomera. Mesoventrite approximately 1.25x as wide as midline length, subacutely projecting at anterior midpoint, uniformly punctate. Metaventrite rather deeply depressed along midline, rather shallowly and sparsely punctate medially, with larger and coarser punctures posterad and laterad, becoming densely and coarsely punctate at sides; metepisternum and metepimeron similarly coarsely and uniformly punctate. First visible abdominal ventrite weakly punctate at middle, more coarsely punctate near metacoxae; visible ventrites 2-4 with single series of rather small punctures, plus all abdominal ventrites crenulately punctate along posterior margins. Meso- and metatrochanters produced at apices; meso- and metafemora punctate on outer surfaces, metafemur more distinctly elongate; mesotibia weakly curved inward, with weak longitudinal carina on outer surface, stronger carina along posterior margin, apex oblique, weakly hooked at inner apex, not obviously produced at outer apex; metatibia more elongate, more weakly curved, and more distinctly produced at outer apex. Tarsi with basal tarsomeres about 1.5x as long as tarsomeres 2-4, apicalmost tarsomeres about twice as long as tarsomeres 2-4.

Male genitalia (Figure 3F): T8 basally deeply emarginate, widened slightly from base to middle, apical margin weakly emarginate; S8 divided medially, sides slightly separated, subtriangular, articulating at basal corners with ventrolateral process of T8; T9 deeply emarginate dorsally, with broad basal apodemes; S9 (spiculum gastrale) with base about one-half width of apex, evenly widened along its length; T10 not subdivided, about twice as broad as long, widest at midpoint, more strongly narrowed to base than to apex, apex shallowly emarginate; basal piece of aedeagus slightly bulbous, about half as long as tegmen, open on left side (not a continuous cylinder), basal foramen opening left; tegmen narrow, subparallel sided in basal half, narrowed slightly toward apex, divided along midline in apical third, apices separate, parallel to tips, apical half curving gradually downward with apices ultimately perpendicular to main tegmen axis; median lobe nearly as long as tegmen, with basal apodemes constituting about half its length, probably extruded dorsally (following Helava (1978)).

Female (Appendix figure 1E): Valvifers broad basally, narrowing slightly beyond midpoint, then expanded to articulation with coxites; coxites rather short, scoop-shaped, quadridentate, with second inner tooth most strongly produced, coarsely punctured on outer and inner surfaces; gonostyle present on inner surface between second and third apical teeth.

Remarks

In the description above, we denote characters based only on endocast specimens in italics, because it is not possible to know exactly how these manifested on the external surface. Given the characters available, this species can be distinguished from other Onthophilus externally by the impunctate lateral areas on the pygidium, and internally by a tegmen that curves rather gradually from the midpoint to the apex, with its apices parallel but slightly separated over about the apical one-third. Phylogenetic analyses place it in a rather isolated position in the genus, without close relatives among extant species.

References

Acknowledgements

We are most grateful to W Etter and O Schmidt (Natural History Museum of Basel) for loaning us the fossils and to D Bajerlein (Adam Mickiewicz University in Poznań) for providing fixed and dried specimens of Onthophilus striatus. We acknowledge B Mähler (Steinmann Institute, Bonn) for triggering this project and thank T Faragó (KIT, Eggenstein-Leopoldshafen) for his help with the reconstruction of tomographic volumes. We also thank T Hörnschemeyer (University of Göttingen) for his help at an early stage of this project and S Legendre (French National Centre for Scientific Research, Paris) for the explanation of the historical Quercy collections. The ANKA Synchrotron Radiation Facility is acknowledged for providing beamtime.

Decision letter

Diethard Tautz, Reviewing editor, Max Planck Institute for Evolutionary Biology, Germany

eLife posts the editorial decision letter and author response on a selection of the published articles (subject to the approval of the authors). An edited version of the letter sent to the authors after peer review is shown, indicating the substantive concerns or comments; minor concerns are not usually shown. Reviewers have the opportunity to discuss the decision before the letter is sent (see review process). Similarly, the author response typically shows only responses to the major concerns raised by the reviewers.

Thank you for submitting your work entitled "Preservation of three-dimensional anatomy in phosphatized fossil arthropods enriches evolutionary inference" for consideration by eLife. Your article has been reviewed by two peer reviewers, one of whom, Russel Garwood, has agreed to reveal his identity. The evaluation has been overseen by Diethard Tautz as the Senior Editor.

The reviewers have discussed the reviews with one another and the editor has drafted this decision to help you prepare a revised submission.

Summary:

The authors show that the internal preservation of the Quercy fossils is unexpectedly excellent, and the authors reveal this with technical skill. The level of preservation is incredible for a non-amber fossil. The preservation of soft anatomy such as an insect's tracheal system makes the study just the kind of thing that should have a large impact in palaeontology. Further, there is actually more behind this manuscript than a technical/techniques paper, and in fact the phylogenetic finding that the Quercy fossil histerid species is most closely related to a North American species is interesting (should mention this in the Abstract). Surely, this biogeographic finding would not have been possible without such remarkable preservation. Regarding the mode of preservation, it is also quite likely that internal decomposition of the original insect was quickly halted because of the relatively antiseptic nature of the phosphate-saturated water that infiltrated the body.

Essential revisions:

One of the reviewers (Russel Garwood) provides direct comments on text passages that should be responded to (appended below). Several of them refer to providing more citations for the general technical background of the study.

Please clarify the following sentences in light of the accompanying comments:

Abstract:

1) “Internal characters in fossil arthropods are crucial to establishing systematic position, ecological role and evolutionary trends.”

I believe this is an oversimplification of a complex issue – see comments below.

Introduction:

2) “Internal characters are crucial for assessing the systematic placement and ecological role of organisms and provide essential information for reliable evolutionary inference. In many arthropods, they include critical systematic characters (Perreau and Tafforeau, 2011).”

This statement is a simplification of what is in fact a complex and under-researched area. Internal characters in fossils are just a form of missing data, and are no more – or less – important, than external data. I suggest the authors check out the following paper: Sansom, R. S. (2014). Bias and sensitivity in the placement of fossil taxa resulting from interpretations of missing data. Systematic biology, syu093.

It’s relatively safe to say that more data is useful in a cladistic phylogenetic context, but (as Sansom 2015 demonstrates) – what is more important when dealing with missing data is to correctly code data as such, rather than introducing false absences (which is the biggest risk when dealing with fossils). Nobody has, to my knowledge, published any work on whether the loss of internal characters leads to biases in phylogenies. This is where it would be a significant issue. I strongly recommend that the authors reword this statement to reflect our current understanding of the impact missing data can have on phylogenetic inference – and indeed put a relevant reference at the end. The current authors correctly cite Perreau and Tafforeau’s (2011) findings, but Perreau and Tafforeau didn’t broach the “reliable evolutionary inference” angle (which is why the current work is novel). For more papers on missing data, I recommend reading these reviews:

Wiens, J. J. (2003). Missing data, incomplete taxa, and phylogenetic accuracy.Systematic Biology, 52(4), 528538.

Wiens, J. J. (2006). Missing data and the design of phylogenetic analyses.Journal of biomedical informatics, 39(1), 3442.

Kearney, M., & Clark, J. M. (2003). Problems due to missing data in phylogenetic analyses including fossils: a critical review. Journal of Vertebrate Paleontology, 23(2), 263274.

3) “Arthropod fossils generally occur as adpressions…” or, indeed, mixtures of these (see e.g. Crato Formation).

4) “(E.g. pyritized in siderite nodules (Nitecki, 1979)).” While there often is a pyrite halo around siderite fossils, the fossils themselves are usually voids, and thus not pyritised see, e.g., any of my (Russell Garwood) Carboniferous arachnid/insect papers. I suggest you correct this (they are also pretty good to CT scan as a result). Some arthropod fossils are indeed pyritised though, I guess Beecher’s trilobite bed (which is also fairly 3D) or the Hunsruck Slate would both be good examples if you want some.

5) “However, in these fossil types, while internal and soft-tissue characters, such as eyes (Duncan and Briggs, 1996) and muscle fibres (Grimaldi, 2009) have been reported (Sutton, 2008), they have been rare and fragmentary.”

Again, this is something of a simplification – a range of soft-tissue, and indeed, internal characters, are known for whole animals (so they are not that fragmentary). They’re also not as rare as the authors suggest. To name some deposits which preserve internal characters (off the top of my head, so biased towards the Palaeozoic as that’s what I know better – but presumably this gets more, rather than less, frequent with younger rocks):

The Rhynie Cherts

Torridonian Cherts

Doushantuo Formation

Orsten

Gogo Formation

La Voulte-sur-Rhône

And all macerates from the Silurian/Devonian (e.g. Gilboa, Wenlock)

If you include just soft tissues, you can add many more, e.g.:

Chengjiang

All Burgess Shale-type deposits

All Ediacaran fossil-bearing deposits

The Silurian Lagerstatten

All Mazon-creek-type Lagerstatten

The Soom Shale

The Hunsrück Slates

Solnhofen Limestone

Crato Formation

I suggest the authors reword to highlight that soft tissue preservation, at least, is surprisingly common, and indeed, that this makes their work more important, as it will apply to a larger number of fossils.

6) Synchrotron X-ray microtomography has become established for detailed examination of fossils (Sutton, 2008). Microtomography has become well established enough in palaeontology that there is a book on it:

Sutton, M., Rahman, I., & Garwood, R. (2013). Techniques for virtual palaeontology. John Wiley & Sons.

The authors shouldn’t feel compelled to cite it, but there are a large number of appropriate references in it that may be of utility. I’d also encourage the authors to just say Microtomography, as palaeontologists have a tendency to run to a synchrotron for any scan they can do, where for the majority a decent lab-based scanner would do the job very well. This is something which doesn’t need to be further entrenched in the mindset.

7) “Recently especially for amber inclusions”

Not especially, as there are a comparable number of applications to fossil plants and seeds in a variety of preservational settings, Doushantuo fossils, macerated microfossils, and many more besides. I recommend the authors say “including fossils preserved in amber” or similar.

8) “Using the same experimental setup”

Why? Surely the beam energy and scanning conditions could (and should) be modified for each specimen?

Results and Discussion:

9) “It is mostly represented by air-filled spaces in the fossil”

What, the cuticle? So you have a void in the form of the organism? This isn’t 100% clear – I suggest you clarify. Furthermore, it’s quite hard to tell preserved cuticle from void in scans, as both will be significantly less dense than the surrounding rock. How can you be sure that features such as, for example, the accessory gland, are not still preserved cuticle.

10) “Large parts of alimentary canal and tracheal system are visible”

This is fantastic.

11) “While the middle part is apparently filled with mineral matrix but well-defined”

This is a bit hard to discern in the image – is this raw, unmodified data? If so, I suggest that adjusting the contrast to better display these features would be absolutely fine, and would make them clearer.

12) “Facilitated an extended description of the species according to modern taxonomic standards (Material and methods)”

Would this description not belong better in the Results, i.e. before this section? This would seem more logical to me.

13) “However, phylogenetic analysis of the more diverse character set now accessible separates these species widely”

This is clear. However, it is worth noting that a cladistic analysis could never show “a direct lineal relationship”. Rather this would be revealed as a sister group relationship between the two taxa, or – more likely – have intermedius as the earliest split in a clade including striatus – intermedius is only actually one node away from this, instead being in a polytomy that is sister to the striatus clade. Maybe you could clarify the above statement to reflect this.

14) “In all most parsimonious solutions O. striatus is nested within a group of mainly Nearctic species related to O. flohri Lewis, 1888.”

Based on what characters? And what characters is O. intermedius placed on the basis of? If these are not both internal then you could have got this without the CT work.

15) “Inclusion of internal characters of male genitalia for O. intermedius yields a better resolved consensus topology and a fuller view of the biogeographic and morphological history of the group.”

What does “better resolved” actually mean? Can you add an SI figure showing the trees of analysis with, and without, genitailic characters? Or with, but with those for O. intermedius coded is unknown in one and correctly coded in the other?

16) “Critical diagnostic differences in external morphology were also revealed by visualization of features previously obscured by matrix.”

What in particular?

This discussion of the phylogeny results could be strengthened by using including the above points. In addition, I had a fiddle with the data in TNT. To make the NEXUS load in this software, it needs to be simplified to:

#NEXUS

BEGIN DATA;

DIMENSIONS NTAX = 37 NCHAR=41;

FORMAT DATATYPE = STANDARD GAP = - MISSING = ? SYMBOLS = " 0 1 2 3 4";

MATRIX

'Cretonthophilus_tuberculatus' 12122143222?12211?312121?1?

'Anapleus_sp.' 1111211144121111113221212111111121211?221

'Peploglyptus_belfragei' 211121222411122111312221111111121?21112?

'Sculptura_kivuensis' 2212124321121221132121212122121?211211?

'Vuatuoxinus_borassicola' 2212222314121211113121212111111?

'Epiechinus_spDNA' 22112243111112211131222221221213221121111

'Epiechinus_rappi' 22122222241112211121212121111112221121112

'Onthophilus_intermedius' ?43211?1221132?121122222211?1211222

'Onthophilus_affinis' 212111211{2 4}2312211322112111111111212211211

'Onthophilus_alternatus' 22222143111212212321222222221222112111222

'Onthophilus_aonoi' 2221114323221222232?22?21111?2?111?1?

'Onthophilus_australis' 22112121141312211?31221111221111112?113?

'Onthophilus_cynomysi' 21212143211212111221221112111112122112322

'Onthophilus_deflectus' 22222143211312221332122111221211212211211

'Onthophilus_flavicornis' 22222143211?2221133?2121?222122?2?21212?

'Onthophilus_flohri' 22222143211312211332122122221221212111211

'Onthophilus_giganteus' 2221224211222221222?21222222?2222212211?

'Onthophilus_globulosus' 22211?432312122113?1?1?22?

'Onthophilus_intermixtus' 222221431113122113322221222212212?21112?

'Onthophilus_irregularis' 222221432313122223?1?22122?

'Onthophilus_julii' 222221432313122223321221?222122?

'Onthophilus_kirni' 21212222241222112121212122221122122112222

'Onthophilus_lecontei' 21212142111212111321211222111112122112222

'Onthophilus_lijiangensis' 2222114323121222232?21212221132?

'Onthophilus_niponensis' 222111432312122223222121?21111?1?111?1?

'Onthophilus_nodatus' 222221432112122123212212122212222111211?

'Onthophilus_ordinarius' 2?432122122223?2?1?2?2?211?2?

'Onthophilus_ostreatus' 22211143232212222322?22?222112?2?111?2?

'Onthophilus_pluricostatus' 222221431112122123212222222212222111213?

'Onthophilus_punctatus' 22211?432312122123?2?2221?2?1?22?

'Onthophilus_sculptilis' 2221114311121222232?222?22211221?1111312

'Onthophilus_silvae' 21222143221212111221112112?122?2?11211?

'Onthophilus_smetanai' 222111?1?1211132?22?1?21111?21112?1?

'Onthophilus_soltaui' 2121114211122211122?2122?221111?

'Onthophilus_striatus' 22222143211312211322122112221121112111211

'Onthophilus_thomomysi' 2121122214122211122?2122221111121?21122?2

'Onthophilus_wenzeli' 22222143111222111321222222221122112212322

END;

I was interested to see where intermedius came out using implied weighting (k=3), shown on the next page. The authors may like to include this in their Discussion, as it tells a slightly different story, and is fairly widely used in fossil arthropod phylogenies. I quote from one of my papers (Garwood and Dunlop, 2014) for more info: “Goloboff (1993) and Goloboff et al. (2008) provide an overview of this weighting scheme, whilst Legg, Sutton & Edgecombe (2013), Legg & Caron (2014) and Ortega-Hernández, Legg & Braddy (2013) provide justification of its use in a palaeontological context.”

Also, just to note, that we have found that viewing external morphology of fossils alone through CT: Garwood, R. J., & Dunlop, J. (2014). Three-dimensional reconstruction and the phylogeny of extinct chelicerate orders. PeerJ, 2, e641.

I would make the case that some of the changes you are seeing are down to just more data, or the inclusion of fossils themselves – as we found in the above exercise. To tie what you are seeing down to the inclusion of internal characters I would like to see a little more interrogation of the data as suggested above – obviously this paper is neither the time nor place to try and quantify this, or even say much in general, but there are relatively simple tests you can do through removal of characters and taxa to try and assess what is driving tree topology.

17) “Based on our examinations we can reconstruct the probable fossilization process of the Quercy Onthophilus specimens.”

I suggest you state here outright that some internals are replaced through mineralisation, and that some cuticle is preserved as void – as it is currently structured it takes quite a long time to get to this fact, which is useful for following the hypothesis for how this was preserved.

18) “By a liquid phosphoric solution”

What kind of solution – phosphate rich water?

19) “From a handful of localities”

If there are relatively few, I suggest you name them so the reader does not have to look them up.

20) “Including fragile appendices”

Do you mean appendages?

21) “Are circulating water”

Virtually any sediment will have circulating water in it. I’d reword to reflect this e.g. “Possible sources for high phosphorous concentrations in water circulating through the fissure fill are…”

22) “In contrast, an artificial ‘digital endocast’”

What you mean by this isn’t 100% clear to me.

23) “So far, the study of the anatomy of fossil arthropods has been largely confined to amber inclusions”

This is not an accurate statement. To choose outline just a limited number of examples of fossil arthropod anatomy in recent years, the Herefordshire group has published >15 papers using serial grinding to elucidate fossil anatomy. Here are three:

Sutton, M. D., Briggs, D. E., Siveter, D. J., Siveter, D. J., & Orr, P. J. (2002). The arthropod Offacolus kingi (Chelicerata) from the Silurian of Herefordshire, England: computer based morphological reconstructions and phylogenetic affinities. Proceedings of the Royal Society of London B: Biological Sciences, 269(1497), 1195-1203.

Briggs, D. E., Sutton, M. D., Siveter, D. J., & Siveter, D. J. (2004). A new phyllocarid (Crustacea: Malacostraca) from the Silurian fossil–Lagerstätte of Herefordshire, UK. Proceedings of the Royal Society of London B: Biological Sciences, 271(1535), 131-138.

Siveter, D. J., Briggs, D. E., Siveter, D. J., Sutton, M. D., & Joomun, S. C. (2013). A Silurian myodocope with preserved soft-parts: cautioning the interpretation of the shell-based ostracod record. Proceedings of the Royal Society of London B: Biological Sciences, 280(1752), 20122664.

Xiaoya Ma and colleagues have published at least four papers on fossil brain anatomy in arthropods:

Ma, X., Hou, X., Edgecombe, G. D., & Strausfeld, N. J. (2012). Complex brain and optic lobes in an early Cambrian arthropod. Nature, 490(7419), 258-261.

Tanaka, G., Hou, X., Ma, X., Edgecombe, G. D., & Strausfeld, N. J. (2013). Chelicerate neural ground pattern in a Cambrian great appendage arthropod.Nature, 502(7471), 364-367.

Cong, P., Ma, X., Hou, X., Edgecombe, G. D., & Strausfeld, N. J. (2014). Brain structure resolves the segmental affinity of anomalocaridid appendages.Nature.

Ma, X., Cong, P., Hou, X., Edgecombe, G. D., & Strausfeld, N. J. (2014). An exceptionally preserved arthropod cardiovascular system from the early Cambrian. Nature communications, 5.

I have published 18 papers since 2010 using CT to describe fossil arthropod anatomy from largely Carboniferous taxa. We’ve even had suggestions of internal anatomy and evo devo in trilobites:

Ortega-Hernández, J., & Brena, C. (2012). Ancestral patterning of tergite formation in a centipede suggests derived mode of trunk segmentation in trilobites. PloS one, 7(12), e52623.

None of which are in amber. The present authors need to rephrase this sentence to reflect the fact they are building on a significant body of work, with a deep history [see Pocock, R. I. (1911). monograph of the terrestrial Carboniferous Arachnida of Great Britain. for some truly excellent work on fossil arthropods anatomy]. There is still a representational bias, but it will be more complex than just arboreal, as it will be defined by taphonomic windows.

24) “May provide a highly complementary source of information on the evolutionary history of arthropods”

Yes, we’ve been saying this for a few years now – here’s a quote from a 2009 paper (Garwood et al 2009 Biology Letters):

“These results demonstrate the ability of XMT to differentiate the void left by the original organism’s decay within sideritic host material and the power of computer-based three-dimensional visualizations of the resultant datasets as a tool for morphological analysis.”

The current work shows the breadth of the technique, and further demonstrates its potential for widespread application. I feel the current authors’ point could be strengthened by citing some of the other 3D work on arthropods, and framed in the light of my comment above. The references don’t have to be my paper – see, for example:

Selden, P. A., Shear, W. A., & Sutton, M. D. (2008). Fossil evidence for the origin of spider spinnerets, and a proposed arachnid order. Proceedings of the National Academy of Sciences, 105(52), 20781-20785.

Bosselaers, J., Dierick, M., Cnudde, V., Masschaele, B., Van Hoorebeke, L., & Jacobs, P. (2010). High-resolution X-ray computed tomography of an extant new Donuea (Araneae: Liocranidae) species in Madagascan copal. Zootaxa, 2427(1), 25-35.)

Materials and methods:

25) “3D X-ray micro-computer tomography”

There is no such thing as computer tomography – I think this should read computed tomography.

26) “Parallel polychromatic X-ray beam, spectrally filtered”

How was the beam filtered? Did you use a filter material? Does the beamline have a variable-period undulator?

27) “The surface polygons were reduced”

Using what algorithm – quadric fidelity reduction? What was the original triangle count, and how much was it reduced by? This will give a useful indicator of just how much smoothing the process will have introduced, and is thus worth mentioning.

DOI: http://dx.doi.org/10.7554/eLife.12129.013

Author response