Pedipalp anatomy of the Australian black rock scorpion, Urodacus manicatus, with implications for functional morphology

Specimen collection

Five individuals of Urodacus manicatus were collected from a private property in Invergowrie, New South Wales, Australia (coordinates: 30°31′29”S, 151°30′20”E). Individuals were sampled at night, located using an ultraviolet (UV) light source during the Australian summer of February 2020. Based on the enlarged mesosoma, relatively small pectines, and lack of genital papillae, the specimens are identified as females; U. manicatus does not exhibit notable pedipalp sexual dimorphism. Two complete specimens, in addition to three pedipalps extracted post-mortem from two individuals are housed in the arthropod collection of the Natural History Museum (NENH-AR) at the University of New England (UNE), Armidale, Australia; specimen numbers are NENH-AR-00015, NEHM-AR-00016, NENH-AR-00017, NEHM-AR-00018, and NENH-AR-00019. The two largest specimens (NENH-AR-00015, NEHM-AR-00016) were initially kept alive in captivity, fed crickets, and used to determine pinch force values. The other two individuals from which pedipalps were removed (represented by NENH-AR-00017, NEHM-AR-00018, NENH-AR-00019) were frozen soon after collection, while NEHM-AR-00015 and NEHM-AR-00016 were frozen after pinch force data were gathered. The largest individual (NEHM-AR-00015) was selected for microcomputed tomography (micro-CT) scanning and reconstruction. The fifth (unregistered) specimen was released after being photographed under sunlight and UV light using a Canon PowerShot SX60 HS camera.

Micro-CT scanning and 3D reconstruction

Specimen NEHM-AR-00015 was micro-CT scanned to document the pedipalp muscles. Prior to scanning, it was submerged in ethanol to dehydrate muscles for more effective identification in the micro-CT scan. A scan with just dehydration through ethanol was conducted. Contrast-enhancement using phosphotungstic acid (PTA) (Metscher 2009; Gignac et al. 2016) was subsequently undertaken with NEHM-AR-00015. In this case, the specimen was immersed in a 70 mL mixture of 70% ethanol and PTA, with a PTA concentration of 0.5%. To examine possible increase in density difference due to the PTA, scans were conducted 12, 36, 48, 72, 168, and 192 hours after submergence. However, these scans were abandoned as they generated lower contrast images compared to ethanol fixation. While this might reflect slow diffusion of the PTA through the specimen, we did not pursue this approach further as the scan of the ethanol-only condition was sufficient. NEHM-AR-00015 was scanned in a GE-Phoenix v|tome|xs micro-CT scanner using the ‘Direct’ tube at UNE, under optimised X-ray tube settings (200 kV, 100 A, 1000 × 1000 pixels capture conditions, an isotropic voxel side length of 22.5 μm, and a rotation step of 0.1125°). The scan focused on the pedipalp and prosoma. Scan data were captured and reconstructed using datos|x software ver. 2.2.1 (phoenix, Wunstorf, Germany).

The scan was imported into Mimics ver. 23.0 (Materialise, Leuven, Belgium) and segmented using the ‘Segmenting’ tool. The prosoma, anterior walking legs, pedipalp podomeres, and pedipalp muscles were segmented using the ‘Segmenting’ tool (Supplemental Fig. 1). Muscle descriptions presented here are based on the scans, muscle fibre directions in scans, 3D reconstructions, and publications previously detailing scorpion pedipalp muscles (Vyas 1970; Shultz 2007). To identify a muscle as an independent unit, two criteria were considered. Firstly, segmented muscles were reconstructed in 3D in Mimics and correspondences to muscles in published examples were identified (Vyas 1970; Shultz 2007). Secondly, other muscle groups were defined by matching morphologies and common attachment points, muscle bands and fibre orientation, and positions in both pedipalps. In taking this approach, we consider the identified muscles to be functionally and morphologically independent. Muscles were described following the terminology in Shultz (2007). This includes the origin location, insertion location, and the morphology of the muscles themselves. Muscles are numbered proximal to distal along the appendage. Once the digital dissection was complete, 3D models of the exoskeleton and individual muscles were exported as .STL files from Mimics and imported into Geomagic Studio (3D Systems, North Carolina, USA). All reconstructions were smoothed in Geomagic Studio. Smoothed .STL files were exported from Geomagic Studio, and a 3D PDF was generated using Tetra4D. Distinct muscles were grouped together and distinctly coloured to delineate the different muscle groups. Supplemental Fig. 1 is available from the following link: 10.17605/OSF.IO/GD7TV. All scan data are available as unedited .TIFF stacks on MorphoSource.org at ark:/87602/m4/495251.

Cuticle microstructure and elemental mapping

The three extracted pedipalps were embedded in epoxy blocks for sectioning in transverse, sagittal, and coronal planes. Embedded appendages were ground through to section the pedipalps. Sectioned surfaces were polished, gold coated and imaged using the JEOL JSM-6010LA Scanning Electron Microscope (SEM) at UNE. Accelerating voltages of 10 kV (NEHM-AR-00019) and 20 kV (NEHM-AR-00017, NEHM-AR-00018) were used to acquire the secondary electron (SE) images of the appendages.

The elemental composition of the chelae teeth and proximal cuticle for NEHM-AR-00018 and NEHM-AR-00019 were derived with SEM Energy Dispersive X-Ray Spectroscopy (EDS). The specimens were gold coated and analysed under 20 kV for 6 h in the JEOL JSM-6010LA SEM to produce a series of elemental maps. Coating arthropod cuticle with inert metals follows previously developed methodologies (Tadayon et al. 2020; Bentov et al. 2021). This approach avoids build-up of charge while specimens are in the SEM. We did not carbon coat the specimens as this would have rendered carbon measurements unreliable. No distinction regarding elemental distributions in EDS has been reported when different coating materials have been used (Bentov et al. 2021). As such, we are confident our approach did not impact the distribution and relative abundance of elements within the examined sections. Additionally, backscatter images of mapped regions were made under these conditions. Preliminary maps were made to reveal the most informative elements for mapping. Elements, such as iron, were therefore not included as they showed no notable abundance in the preliminary maps.

Chelae force measurements

To measure the pinch force, a force-sensitive resistor (FlexiForce A101 Sensor; Tekscan Inc., Massachusetts, United States) was constructed using a voltage divider circuit (Fig. 2a). The A101 sensor was used as its small size (3.8 mm diameter; Fig. 2d) was the most appropriate for measuring pinch force of U. manicatus pedipalps, and helped to avoid unequal loading during pinching. A 50 Hz frequency was used with an analogue input pin and an Adafruit Feather ATSAMD21 Cortex M0 Adafruit (Adafruit Industries, New York, United States) prototyping board. This approach to building a force transducer had not previously been conducted and was novel for this project (Fig. 2c). Sensor calibrations were needed to calculate pinch force. These calibrations were derived by adding weight to the sensor, increasing the force registered through gravity. Two linear equations representing the logarithmic relationship between resistance and force were derived using Microsoft Excel and used to calculate force based on the measured values, while allowing the closest fit of data to calibration values (Fig. 2b; Supplemental Data 1). This approach allowed force data to be gathered at a frequency of 1000 Hz. However, to limit the dataset to a manageable number of observations (<1000), we gathered data at a frequency of 50–100 Hz. To derive pinch force values, NENH-AR-00015 and NEHM-AR-00016 were agitated and prompted to pinch down on the resistor. The outer edge of the force transducer was positioned within the pincer, close to the moveable and fixed finger articulation to induce pinching and avoid unequal loading. This experiment was conducted between 20 and 25°C. This resulted in a dataset of 836 distinct force data points from seven distinct pinch events (NEHM-AR-00015) and 788 distinct force data points from four distinct pinch events (NEHM-AR-00016). Data were plotted as histograms to examine pinch force distribution.

Fig. 2.

Details of force transducer for pedipalp pinch force tests. (a) Schematic of the circuit used for the transducer. (b) Calibration data used to derive equations for determining force output from the transducer. (c) An image of the force transducer. (d) Example of force plate used. Abbreviations: A0, Analogue measuring point; FSR, Force sensitive resistor; N, Newtons; VCC, Power source (5 V); 10k, Voltage dividing resistor.

Pedipalp musculature

The anteriormost walking legs, fixed and movable cheliceral segments, labrum, prosoma, and pedipalp podomeres (coxa, trochanter, femur, patella, tibia, and tarsus) were resolved (Fig. 3). The pedipalp morphology consists of reduced, stout coxae and trochanters, elongated femurs and patellae, bulbus manus regions in the tibia, with elongate and robust tibial and tarsal fingers. In total, 25 muscle groups associated with the pedipalps were identified (Table 1; Figs 4, 5, 6, 7 and 8), expanding on the muscles presented by Vyas (1970) and Shultz (2007) (Table 1). These additional muscles are primarily located within the trochanter (Muscles 8–13) and are generally smaller and located proximally to larger muscle groups, so may have been missed or grouped together in previous studies (Table 1). The 25 muscles include groups with a prosomal origin (Muscles 1–5; see depiction of Muscles P1, P2, P3, P8, P9 in Shultz 2007; Table 1) and groups located wholly within the pedipalps (Muscles 6–25). Most pedipalp muscles have an origin proximally within their respective appendage segments (Figs 4, 5, 6, 7 and 8) and insert within or proximal to the next distal podomere (Table 1). There are two muscles with large tendons that extend to more distal podomeres: Muscles 9 and 24. In the micro-CT scan, several longer muscle groups show breakage or truncation. In particular, we note that Muscles 1, 2, and 21 and the tendons of Muscles 9 and 24 show this condition. These breakages reflect fixing from ethanol and muscle damage during dehydration.

Fig. 3.

3D reconstructions of the anterior region of Urodacus manicatus, as modelled by micro-CT scanning. NEHM-AR-00015. (a) Dorsal view. (b) Ventral view. (c) Anterior view excluding patella, tibia, and tarsus. (d) Posterior view excluding patella, tibia, and tarsus. The 3D PDF is found in Supplemental Fig. 1.

Fig. 4.

3D reconstructions of all identified prosomal and pedipalp muscles in NEHM-AR-00015, as modelled from micro-CT scanning. No key is included as some groups are not easily visualised when all muscles are illustrated. Separate muscles are illustrated in Figs 5–8. (a, c, e, g) Muscles with the exoskeleton as transparent. (b, d, f, h) Muscles without the exoskeleton. (a, b) Dorsal view. (c, d) Ventral view. (e, f) Anterior view. (g, h) Posterior view. All muscles are coloured separately. The 3D PDF associated with this reconstruction is found in Supplemental Fig. 1.

Fig. 5.

Pedipalp muscles with prosomal origins, as modelled by micro-CT scanning. NEHM-AR-00015. (a, b, e) Prosoma and coxa as transparent. (c, d, f) Muscles without the exoskeleton. (a, c) Dorsal view. (b, d) Ventral view. (e, f) Left lateral view. Abbreviations: co, coxa; pro, prosoma. The 3D PDF associated with this reconstruction is found in Supplemental Fig. 1.

Fig. 6.

Pedipalp muscles with coxal origins, as modelled from micro-CT scanning. NEHM-AR-00015. (a, c) Coxa, trochanter, and femur transparent with muscles. (b, d–j) Muscles without the exoskeleton. (a, b) Dorsal view. (c, d) Ventral view. (e) Anterior view. (f) Posterior view. (g–j) Left lateral view showing digital dissection of superimposed muscles. (k) Posterior view owing digital dissection of superimposed muscles. Abbreviations: co: coxa; fem: femur; tro: trochanter. The 3D PDF associated with this reconstruction is found in Supplemental Fig. 1.

Fig. 7.

Pedipalp muscles of the trochanter and femur, as modelled from micro-CT scanning. NEHM-AR-00015. (a, c, e, g) Trochanter and femur transparent with muscles. (b, d, f, h–m) Muscles without the exoskeleton. (a, b) Dorsal view. (c, d) Ventral view. (e, f) Anterior view. (g, h) Posterior view. (i, j) Ventral view of right appendage showing digital dissection of superimposed muscles. (k, l) Dorsal view of right appendage showing digital dissection of superimposed muscles. (m) Anterior view of left appendage showing digital removal of Muscles 14 and 17. Abbreviations: fem: femur; tro: trochanter. The 3D PDF associated with this reconstruction is found in Supplemental Fig. 1.

Fig. 8.

Pedipalp muscles from the patella, tibia, and tarsus, as modelled from micro-CT scanning. NEHM-AR-00015. (a, c, e) Patella, tibia, and tarsus transparent with muscles. (b, d, f) Patella, tibia, and tarsus muscles with transparent tarsus. (a, b) Dorsal view. (c, d) Ventral view. (e, f) Posterior view. (g–j) Muscles 24 and 25 without the exoskeleton showing Muscle 25 as transparent. Dashed white line indicates the direction of Muscle 24. (g, h) Left appendage. (i, j) Right appendage. (g, i) Lateral view. (h, j) Medial view. Abbreviations: pat: patella; tar: tarsus; tib: tibia. The 3D PDF associated with this reconstruction is found in Supplemental Fig. 1.

Pedipalp microstructure

Under the SEM, cuticle structures (sensuRichards 1951; Dalingwater 1987) were observed along the sectioned appendages (Fig. 9). In the sagittal section (NENH-AR-00017), laminate endocuticle is ~70 μm thick, with single lamina ~5 μm thick, and exocuticle 8 μm thick (Fig. 9d). There is a notable lack of laminae in the tibial fixed finger, but 14–60 μm wide pore canals are observed here (Fig. 9b).

Fig. 9.

SEM-SE images of a sagittal section of the pedipalp. NEHM-AR-00017. (a) Section through the femor, patella, and tibia regions; boxes outline the three areas illustrated in b–d. (b) Section of fixed tibia finger. Note limited laminate cuticle and extensive pore canals (white arrows). (c) Close-up of articulation between patella and femur showing laminate endocuticle. (d) Close-up of the proximal tibia edge showing laminate endocuticle (white line) and thin exocuticle (black line). The white outermost layer is epoxy resin, not cuticle. Abbreviations: end: endocuticle; exo: exocuticle; fem: femur; pat: patella; tib: tibia.

Cuticle morphology along both fingers is visible in the coronal section (Fig. 10a; NENH-AR-00018). Laminate endocuticle (~80–200 μm, single lamina ~5 μm thick) is overlain by 70–130 μm-thick mesocuticle (sensuZhang et al. 2023) that lacks any structure when viewed coronally. Above the mesocuticle is thin exocuticle (12–17 μm) with 2–5 laminae (~4 μm thick) (Fig. 10a). Pore canals up to 180 μm wide are present in the finger and proximal tarsal regions. Finger teeth are darker than other cuticular areas in backscatter imagery and show tooth roots extending into the fingers (Fig. 10b).

Fig. 10.

SEM images and EDS elemental maps of coronal section of chela. NEHM-AR-00018. (a) SEM image of sectioned tibia and tarsus showing pore canals in the moveable and fixed fingers. Lines show cuticle section thickness. Box outlines close up in (b–m). (b) Backscatter image of box in (a) showing teeth (black arrows). (c–m) Elemental maps of zinc, sodium, chlorine, phosphorous, potassium, vanadium, calcium, oxygen, nickel, magnesium, and carbon, respectively. Note the teeth roots in (c–g). Scale bar in (b) is the same for (c–m). Abbreviations: end: endocuticle; exo: exocuticle; mes: mesocuticle; tib: tibia; tar: tarsus.

In the transverse section (Fig. 11; NENH-AR-00019), the laminate endocuticle is ~90 μm thick, with single lamina ~5 μm thick. Regions where the endocuticle extends to the teeth are observed, with a thickness of ~290 μm (Fig. 11a). Where the endocuticle does not extend to the teeth, the endocuticle is overlain by mesocuticle, consisting of vertically stacked cuticle that is 60–70 μm thick. The mesocuticle is overlain by exocuticle that is 20–30 μm thick, with ~4 μm thick laminae. Pore canals reaching 165 μm in length and cross-cut endo-, meso-, and exocuticle layers. Teeth are lighter than other cuticle in backscatter imagery (Fig. 11b).

Fig. 11.

SEM images and EDS elemental maps of transverse section of chela. NEHM-AR-00019. (a) SEM image of sectioned tibia and tarsus fingers showing endocuticle, mesocuticle, and pore canals. Box outlines close up in (b–m). (b) Backscatter image of box in (a) showing teeth (white arrows). (c–m) Elemental maps of zinc, sodium, chlorine, phosphorous, potassium, vanadium, calcium, oxygen, nickel, magnesium, and carbon, respectively. Scale bar in (b) is the same for (c–m). Abbreviations: end, endocuticle; exo, exocuticle; mes, mesocuticle; tib, tibia; tar, tarsus.

Elemental construction of chelae cuticle and teeth

Elemental maps of the coronal and transverse sections focusing on the teeth show similar elemental distributions (Figs 10, 11; NENH-AR-00018, NENH-AR-00019). There is a pervasive enrichment of chlorine (Cl), nickel (Ni), phosphorus (P), potassium (K), sodium (Na), vanadium (V), zinc (Zn), and oxygen (O) in the teeth and roots (Figs 10, 11) and a notable lack of carbon (C) in the teeth. Calcium (Ca) has limited enrichment along the chelae edges and teeth roots. Magnesium (Mg) has a consistent distribution across teeth and cuticle. The general pedipalp cuticle is distinctly lacking in Na, Zn, V, and Ni, has a limited abundance of Cl, P, K, Ca, and O, and an enrichment of C relative to teeth.

Pedipalp pinch force

Pinch force values from specimens NENH-AR-00015 (n = 836 measurements) and NENH-AR-00016 (n = 788 measurements) have similar distributions. Both distributions are right skewed, with major peaks at about 1 N (Fig. 12). NENH-AR-00016 has two smaller peaks at 2.8 N and 4.2 N, and a mean pinch force of 1.84 N. NENH-AR-00015 has three smaller peaks at 2.0 N, 3.2 N, and 4.2 N, and a mean pinch force of 2.22 N. NENH-AR-00015, the larger of the two studied individuals, tended to produce higher pinch forces.

Fig. 12.

Histogram plots of pinch force values for Urodacus manicatus. Data associated with these plots are found in Supplemental Data 2.

Using micro-CT to inform on scorpion anatomy

Our documentation of pedipalp muscles expands on previous research by Vyas (1970, on Heterometrus fulvipes) and Shultz (2007, on Heterometrus spinifer) by presenting 20 muscle groups located solely within pedipalp podomeres and five muscle groups with prosomal origins. Based on the origin, insertion, and muscle shape, we can suggest homologies between 18 of these muscles across these studies (Vyas 1970; Shultz 2007). This demonstrates that at least seven muscles documented herein are novel. These data illustrate how pedipalp muscles engage distinct podomeres. For example, the long tendon of Muscle 9 illustrates that pedipalp coxal muscles engage more distal pedipalp regions than previously realised. These muscle reconstructions have the potential to support arguments about possible homologies between the pedipalp and walking leg muscles of scorpions (see Shultz 2007 for a detailed discussion). In comparing muscles reconstructed here with previous depictions of walking leg muscles (Snodgrass 1952; Shultz 1989; Wolf and Harzsch 2002), similarities in muscle arrangements in the femur and patella can be drawn. In particular, we propose that pedipalp Muscles 17 and 18 of Urodacus manicatus (Fig. 7) are homologous with leg Muscles 19 and 16, respectively, of Pandinus sp. illustrated by Snodgrass (1952, fig. 19G), and Muscles 22 and 23 (Fig. 7) correspond to Muscles 21 and 20 (cf. Snodgrass 1952, fig. 19G). A detailed examination of these muscles using similar tools is needed to document these proposed homologies more thoroughly and confidently.

Functional interpretations for musculature can be drawn using muscle origins, insertions, and size. Muscles 1–5 have prosomal origins, inserting into the coxae. As they are extrinsic muscles (Beck 1885; Lankester 1885; Shultz 2007), they reinforce and maintain the coxal position. Muscles 6–8 and 10–13 are positioned within the coxae, with insertion into the trochanter, likely permitting the large range of motion for the articulation, similar to muscles in the walking legs (Snodgrass 1952). Muscle 9, extending from the coxa to the patella as a tendon, may function to rapidly retract the more distal pedipalp podomeres, possibly during or after capturing prey. This muscle apparently does not have an immediate analogue for comparison within other scorpion appendages. However, Muscle 9 is morphologically comparable to Muscle 10 of horseshoe crab walking legs (Snodgrass 1952, fig. 11a; Bicknell et al. 2018a, fig. 6) which is a flexor that moves the patella and other distal podomeres of the xiphosurid appendage towards more proximal podomeres and the body (Shultz 1989; 2001). A similar tendon extends across multiple appendage podomeres in the scorpion pedipalp. This morphological similarity suggests a comparable function – as a flexor muscle that moves the appendage towards the body, after attacking, or in defence. Such an explanation is also possible for Muscle 21 that could be used to retract the patella. Muscles 14–16 inserting proximal to the patella from the femur also permit a large range of motion for the patella–femur articulation. Muscles 17–20 are likely used to rotate and position the pedipalps relative to the prosoma. Developing on the homology with the walking leg muscles documented by Snodgrass (1952), Muscles 22 and 23 are flexor and extensor muscles, respectively. Among other possible uses, such as mating (Kellersztein et al. 2019), the large size of these muscles can be used to pull prey to the mouth for processing. Finally, Muscles 24 and 25 are well documented in other taxa as the muscles used in the closing and opening of the tarsal podomere (Snodgrass 1952; van der Meijden et al. 2012a; Simone and van der Meijden 2021). More confident determination of the muscle roles discussed here requires in vitro examination of muscle movement (see Walker et al. 2014).

Identifying and reconstructing muscle clusters in 3D represents an important approach for testing hypotheses on the synapomorphies in related groups. One possible direction for this application of 3D anatomy is the grouping of scorpions and pseudoscorpions into Panscorpiones based on phylogenomics and rare genomic changes (Ontano et al. 2021); the traditional sister group hypothesis between pseudoscorpions and solifuges is challenged by the latter not sharing a whole genome duplication characteristic of arachnopulmonate arachnids (Gainett et al. 2024). In this case, one could determine whether the muscles groups in scorpion pedipalps are also observed in pseudoscorpion taxa. While pseudoscorpion coxal muscles have been documented in 3D (Michalski et al. 2022), a detailed consideration of muscles in their pedipalps has not yet been undertaken in a 3D context.

Documenting pedipalp muscles of other scorpion species in 3D to build on our data may be useful for uncovering possible anatomical conservation in scorpions. Examining metasomal muscles of different scorpion genera in this framework has shown how modifying metasomal segment morphologies impacts muscle shape, but conserves muscle groups (Günther et al. 2021a, 2021b). A similar situation is expected for pedipalp muscles. However, as scorpion chelae are extremely morphologically diverse (van der Meijden et al. 2010, 2012a; Simone and van der Meijden 2018; Bicknell et al. 2022), modifications to distal pedipalp muscles are likely. Such muscle data represent as-yet unexplored ecological and phylogenetic signals.

Microstructure of the chelae

Our observations of the cuticular microstructure align with previous research (Mutvei 1978; Kellersztein et al. 2019; Zhang et al. 2023), demonstrating common cuticle construction across scorpion chelae. Among the microstructures we observed, the dorsoventrally oriented mesocuticle laminae has been noted only within the chelae and walking legs (Zhang et al. 2023). This modified cuticle provides reinforcement by orienting cuticle in the same direction as the maximum force, similar to the vertically oriented cuticle in Limulus polyphemus (Linnaeus 1758) gnathobases (Bicknell et al. 2018b). This increases chelae robustness overall.

Mechanical and functional interpretations can be derived by considering the three main cuticle layers. The presence of these layers in the coronal and transverse orientations demonstrates how the Urodacus manicatus chela cuticle is constructed to experience higher stresses and strains (Kellersztein et al. 2019). The marked thickness of the endo- and mesocuticle, compared to the exocuticle, permits more plastic deformation within the structure, allowing for higher loads to be tolerated (Zhang et al. 2023). The thick endocuticle in the sagittal orientation illustrates that chelae fingers are reinforced to limit development of cracks within the cuticle (Zhang et al. 2023). These observations illustrate that U. manicatus chelae are well adapted for handling prey (Farley 1999; Simone and van der Meijden 2018; Kellersztein et al. 2019).

Examination of cuticle microstructure here failed to differentiate exocuticle into hyaline and inner exocuticle regions – the two main divisions of scorpion exocuticle (Filshie and Hadley 1979; Dalingwater 1987; Rubin et al. 2017). This likely results from our study of unstained specimens. Staining, such as Mallory’s Triple Stain technique (Mallory 1900) involves the use of heavy-metal acids (e.g. phosphotungstic acid). We avoided this technique, as it would have impacted the results of the elemental mapping. However, the florescence of Urodacus manicatus under UV light (Fig. 1b) confirms the expected presence of hyaline exocuticle (Rubin et al. 2017).

Elemental composition of the chelae

Elemental mapping revealed several heavy elements in the cuticle. Many arthropods utilise heavy elements to reinforce cuticular structures used in food processing, such as mandibles and raptorial appendages (Hillerton et al. 1984; Fontaine et al. 1991; Schofield et al. 2002, 2021; Lichtenegger et al. 2003; Morgan et al. 2003; Cribb et al. 2008; Gallant and Hochberg 2017; Bentov et al. 2021). Zinc functions as a cross-linker of different proteins to increase hardness and provide reinforcement (Lichtenegger et al. 2003; Cribb et al. 2008, 2010; Gallant et al. 2016), allowing for higher mechanical loads (Tadayon et al. 2020). Increased zinc abundance in arachnids is thus far limited to chelicerae and pedipalp regions (Schofield et al. 1989; Schofield 2001; Cutler and McCutchen 2006; Politi et al. 2012; Gallant et al. 2016; Gallant and Hochberg 2017; Radosavljevic et al. 2021) and is associated with cuticle hardening for handling prey and interacting with substrate (Radosavljevic et al. 2021). Here, the increased zinc in Urodacus manicatus chelate teeth demonstrates similarly robust cuticle. A comparable abundance of chlorine in teeth reflects the presence of chlorides bonded to zinc (Hillerton and Vincent 1982; Schofield 2001; Politi et al. 2012) – a condition already known in scorpions (Schofield et al. 2003; Radosavljevic et al. 2021). The lack of iron within the pedipalps, as observed in the initial scans, reflects a similar pattern of enriched zinc and paucity of iron in non-buthid scorpions (Schofield 2001). The other very abundant element was sodium. However, as NaK and ZnL shell energies overlap (Radosavljevic et al. 2021), the presence of sodium records the overlap of energy shells, not a true abundance. Finally, nickel, potassium, and vanadium may reflect similar reinforcement as zinc in the teeth (Gallant et al. 2016), but their lower abundances likely demonstrate limited involvement in the hardening process.

Phosphorus is uncommon within terrestrial chelicerate cuticle (Cutler and McCutchen 2006; Gallant et al. 2016). When present, phosphorus is incorporated as calcium phosphate [Ca₃(PO₄)₂] (Currey et al. 1982; Becker et al. 2005; Gallant et al. 2016). As we observed phosphorus and calcium in the teeth and roots (Fig. 10f, i), the presence of calcium phosphate seems likely. An alternative form of calcium in the cuticle is calcium oxalate (CaC₂O₄) (Norton and Behan-Pelletier 1991; Gallant and Hochberg 2017). However, the lack of carbon in roots and teeth (Figs 10m, 11m) excludes this possibility. Calcium has been documented only in spider and whip scorpion exoskeletons (Schofield et al. 2003; Radosavljevic et al. 2021). The record here suggests that calcium is likely more abundant in the chelicerate exoskeleton than previously realised.

The pathway for incorporating additional elements into exoskeletal structures is worth considering, as these elements are very uncommon within other exoskeletal regions. Heavier and otherwise uncommon elements are obtained through diet (Vohland et al. 2003; Gallant and Hochberg 2017) or the environment (Schofield 2001) and incorporated into the exocuticle either during sclerotisation (Cribb et al. 2010) or, in the case of zinc, after sclerotisation (Schofield et al. 2002). The process of incorporating these elements into the teeth is likely related to the roots – these structures are defined by an abundance of zinc, sodium, phosphorus, and potassium (Figs 10c, d, f, g and 11c, d, f, g), with limited vanadium and calcium (Fig. 10h, i). It seems likely that tooth roots allow zinc and other elements to move through the cuticle after sclerotisation (Schofield et al. 2003).

The elemental distributions demonstrate how teeth of Urodacus manicatus chelae are reinforced with different elements and compounds. Metal elements such as zinc, nickel, potassium, and vanadium within the teeth illustrate that they are present to decrease wear and increase the resistance to localised force. This aligns with the three main cuticle layers oriented for more plastic deformation to limit cracks developing within the cuticle during use. Taken together, the teeth and surrounding cuticular region on the chelae are reinforced for functions such as burrowing, mating, or prey capture (Kellersztein et al. 2021).

Most research considering the elemental construction of arthropod cuticle has employed spectrograms of specified regions (e.g. Cribb et al. 2008; Gallant et al. 2016; Gallant and Hochberg 2017; Radosavljevic et al. 2021). This approach uncovers an overall chemical signal but fails to demonstrate elemental distribution. By mapping elements using EDS, a thorough depiction of elemental concentrations in different regions can be determined, and structures such as the teeth roots can be detected (see also Schofield et al. 1989, 2002; Fawke et al. 1997; Politi et al. 2012; Tadayon et al. 2020; Bentov et al. 2021). While EDS mapping is more time-consuming than spectrograms, it presents a more detailed illustration of element location.

Urodacus manicatus pinch force and ecology

Scorpion chelae have disparate functions, the most common of which are burrowing, mating, prey capture, and occasional prey termination (van der Meijden et al. 2012a; Kellersztein et al. 2021; Bicknell et al. 2022). The strongest Urodacus manicatus chelae pinch force values (4.16 N, Supplemental Data 2) are comparable to those of the African fat-tail scorpion [Androctonus amoreuxi (Audouin 1826)] (4.4 N) and the yellow forest scorpion [Opistophthalmus boehmi (Kraepelin 1896)] (4.4 N) (Bicknell et al. 2022; Supplemental Table 3), both of which have chelae used for defence and predation (van der Meijden et al. 2010, 2013), although we acknowledge that comparison is complicated by differences in animal size. Combined with the robust morphology and the array of adaptions within the cuticular microstructure and elemental composition, this high pinch force suggests that U. manicatus likely uses its chelae to attack, capture, and crush prey. This contrasts with forms that display lower force values (<1 N) and rapid chelae closing speeds, using the chelae to hold prey for incapacitation by the stinger (Casper 1985; van der Meijden et al. 2013; Simone and van der Meijden 2018). This observation also aligns with the life mode of U. manicatus as a predator that waits at burrow entrances for prey (Holden 1997) and uses pedipalps, rather than the stinger, to initially subdue prey (Simone and van der Meijden 2018).

A major complication in attempting to more thoroughly understand scorpion feeding ecology is the infrequent consumption of prey (Simone and van der Meijden 2021). Urodacus manicatus, in particular, can survive for up to eight months without food (Southcott 1955). As such, our morphofunctional interpretations of predation are not supported by behavioural observations. However, the application of chelae to subdue prey aligns with the substantial reinforcement of chelate teeth with metals, the additional mesocuticle, and a large tarsal muscle. Field studies of this and other scorpion species, supplemented by metabarcoding of gut contents (Simone et al. 2022), are required to validate these proposed ecological hypotheses.

Synthesis of methodologies

The examination of one biological structure using four methods has given insight into the morphofunctionality of Urodacus manicatus and, more broadly, offers an integrated framework for exploring a key predatory adaptation. Pedipalp musculature is more elaborate than previously noted, including multiple muscles used in prey capture and manipulation. The microstructural data show how thick chelae endo- and mesocuticle allow for plastic deformation, while promoting robustness for enduring increased force during predation. This is supported by the high concentration of metals within the chelae teeth that likely reduce the wear these regions experience. Finally, the force data allow us to contextualise the mechanical performance compared to other species, while ground-truthing the microstructural and skeletomuscular data. Together, these adaptions and morphologies result in a biological structure employed in prey capture and processing.