Zoological Journal of the Linnean Society, 2023, XX, 1–15 https://doi.org/10.1093/zoolinnean/zlad112 Advance access publication 23 September 2023 Original Article Original Article Bone histology of the Late Pleistocene Prolagus sardus Downloaded from https://academic.oup.com/zoolinnean/advance-article/doi/10.1093/zoolinnean/zlad112/7281581 by guest on 27 September 2023 (Lagomorpha: Mammalia) provides further insights into life- history strategy of insular giant small mammals Eva Fernández-Bejarano1, Alejandro Blanco1, , Chiara Angelone2–4, Zhaoqun Zhang3,5, Blanca Moncunill-Solé1,2,*, Departamento de Física e Ciencias da Terra, Universidade da Coruña, 15008 A Coruña, Spain 1 2 Dipartimento di Scienze, Università degli Studi Roma Tre, 00146 Roma, Italy 3 Key Laboratory of Vertebrate Evolution and Human Origins, Institute of Vertebrate Paleontology and Paleoanthropology, Chinese Academy of Sciences, 100044 Beijing, China 4 Institut Català de Paleontologia Miquel Crusafont, Campus de la Universitat Autònoma de Barcelona, 08193 Cerdanyola del Vallès, Spain 5 College of Earth and Planetary Sciences, University of Chinese Academy of Sciences, Beijing 100049, China * Corresponding author. Departamento de Física e Ciencias da Terra, Universidade da Coruña, Campus da Zapateira, Rúa da Fraga 10, 15008 A Coruña, Spain. E-mail: blanca.moncunill@udc.es, blanca.moncunill@gmail.com ABSTR ACT Fossils provide an excellent opportunity to study and understand the evolution of insular environments free of human-made perturbations. Here, we evaluated the life-history traits and strategy of the extinct insular giant Prolagus sardus (Mammalia: Lagomorpha) by examining microscop- ically its fossilized bone tissues (osteohistology, skeletochronology, and quantitative geometry). For this task, a complete ontogenetic series of femora retrieved from the Late Pleistocene Grotta della Medusa (NW Sardinia, Italy) were analysed. Our results reveal that: (i) P. sardus’ pups were weaned at large size; (ii) the species’ maturation was delayed in its life cycle; and (iii) P. sardus lived longer than expected for its size. Hence, the giant P. sardus should have a slow pace of life triggered by the low levels of extrinsic mortality of the insular habitat. On the other hand, bone tissue differences between P. sardus and Ochotona were found in the vascularization, slow- growing bone presence, and cortical and medullary growth trajectories. The results obtained in this study concur with the eco-evolutionary responses described to date in extinct insular lago- morphs, as well as provide new empirical evidence about the phenomenon known as ‘insular gigantism’ as a pattern of evolution of small-sized mammals in genuine insular ecosystems. Keywords: insular gigantism; Grotta della Medusa; life-history strategy; Lagomorpha; Mammalia; palaeohistology; osteohistology; Sardinia; skeletochronology INTRODUCTION Smith 2005, Alves and Hackländer 2008). However, in some Lagomorphs are key species in ecosystems, i.e. they are the basal cases their dispersal to new regions was not ‘natural’ and oc- trophic level of some food webs, affect the hydrology and soil curred as a consequence of human-caused introductions (acci- biology, model plant communities, etc. The Order Lagomorpha dental or intentional, e.g. Oryctolagus and Lepus because of their comprises 91 extant species of two families, Ochotonidae and significance for human activities) (Chapman and Flux 1990, Leporidae. Ochotonidae consist of a single extant genus, 30 spe- Bonino et al. 2010 and references therein; Fenner 2010). The cies with a clearly relictual distribution in confined areas of cen- success and dominance of leporids in present-day ecosystems tral Asia and North America; whereas Leporidae have 61 species was established only a relatively short time after their dispersal pertaining to 11 genera, some of them monotypic, showing a from North America to Eurasia, via Berginia, which occurred worldwide distribution (excepting Antarctica) and a more vari- during the Late Miocene (c. 8 Mya), and later to Africa (Flynn able range of adaptations and ecological roles (Hoffmann and et al. 2014). The leporid genera that gave rise to the Eurasian Received 18 May 2023; accepted 31 July 2023 © 2023 The Linnean Society of London. This is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs licence (https://creativecommons.org/licenses/ by-nc-nd/4.0/), which permits non-commercial reproduction and distribution of the work, in any medium, provided the original work is not altered or transformed in any way, and that the work is properly cited. For commercial re-use, please contact journals.permissions@oup.com 2 • Fernández-Bejarano et al. ‘invasion’ had thrived confined in North America since c. 15 Mya. Roma, Italy). For comparison purposes, we used 13 femora of Prior to the Miocene, the taxonomic and palaeobiogeographic three extant ochotonid species (Table 1) that recent studies scenario was completely different: Ochotonidae and other ex- highlight as the closest extant relatives to Prolagus, which prob- tinct lagomorph suprageneric, formal or informal, entities (i.e. ably represents a familiar rank of its own [see Family Prolagidae Prolagidae and stem lagomorphs) were incredibly diverse, abun- in Gureev (1960) and Bover et al. (2022), with references]: dant, widespread, and, in some cases, extremely long-lasting Ochotona collaris (Nelson, 1893) (N = 5), Ochotona princeps (López-Martínez 2008, Angelone et al. 2019). (Richardson, 1828) (N = 3), and Ochotona dauurica (Pallas, Downloaded from https://academic.oup.com/zoolinnean/advance-article/doi/10.1093/zoolinnean/zlad112/7281581 by guest on 27 September 2023 Recent studies were dedicated to investigating a possible 1776) (N = 5). The first two species belong to the talus-ecotype causal effect of climate change on the palaeodistribution and (non-burrowing leapers that dwell in rock piles), whereas the palaeobiodiversity of some lagomorph taxa with, interestingly, latter is typical of the meadow-ecotype (burrowers that inhabit non-univocal results. This re-opened the debate whether the ef- meadow and steppe ecosystems) (Smith 1988). The bones of Oc. fects can be attributed solely to climate change vs. (or in addition princeps and Oc. collaris are curated at the Museum of the North to) palaeogeography plus other biological aspects (Angelone and Mammal Collection of the University of Alaska Museum (UAM, Zhang 2021, Čermák et al. 2021). In this respect, Prolagus sardus Fairbanks, USA) and those of Oc. dauurica belong to the collec- (Wagner, 1829), the last representative of a 24 Myr long-lasting tion of the Museum of Southwestern Biology of the University European lineage and the sole taxon other than Ochotonidae of New Mexico (MSB, Albuquerque, USA). and Leporidae to have survived to historical times (Angelone Extant lagomorphs show minimal sexual dimorphism (Lu et al. 2019 with references), represents a unique window to the et al. 2014, Ferreira et al. 2015), so sex is not considered as a past. Its study can provide clues to understanding the response discriminant feature in the present study. We categorized the of lagomorphs to changing environments (particularly to insu- specimens into three age classes, on the basis of growth plate larity), which is useful in the framework of current natural and fusion: (i) juvenile ( J) (= with unfused epiphyses or one fused anthropically driven changes of the present-day world (Leack et epiphysis though with visible suture lines), (2) young adults (Y) al. 2015). (= with fused epiphyses but some of them still show visible su- The study of biological hard structures of past species at the ture line); and (3) adults (A) (= with both epiphyses fully fused tissue level provides meaningful details about their biology (e.g. and suture lines not distinguishable) (Kauhala and Soveri 2001, life-history traits, growth strategies, etc.), which are impossible Moncunill-Solé et al. in press). In broken specimens (N = 2), to achieve with traditional anatomical approaches (de Buffrénil the age category was refined according to metrics and histology. et al. 2021a). In the last decade, several investigations using this The measurements (in mm) were taken using a digital calliper microscopic technique have successfully provided valuable bio- (error ± 0.05 mm) following standards (Moncunill-Solé et al. logical insights into some extinct lagomorphs (Kolb et al. 2015, 2015: fig. 1): anteroposterior diameter at the midshaft (DAPm), Moncunill-Solé et al. 2016). The present research provides a transversal diameter at the midshaft (DTm), maximum fem- histological analysis of the P. sardus remains, considering their oral length (FLmax), femoral proximal transversal diameter ontogenetic variability, thanks to the abundant remains retrieved (FTDp), femoral distal anteroposterior diameter (FAPDd), and from Late Pleistocene Grotta della Medusa (NW Sardinia, Italy; femoral distal transversal diameter (FTDd). 22.390–21.910 ka cal BP; see Palombo et al. 2017). Our analyses The body mass (BM) of adult individuals of P. sardus was aim to improve the interpretations and conclusions provided reconstructed using predictive regression equations. BM re- previously (that are limited by samples not covering all the onto- construction models were computed based on a large set of ex- genetic categories) and to give key details to understanding the tant lagomorph species (N = 44), with worldwide distribution biological strategies adopted by small mammals under insular and covering the size diversity of Lagomorpha (for details see selective pressures in habitats that have not been perturbed by Moncunill-Solé et al., 2015). The BM mean was calculated by human action. averaging the weight estimation using each of the epiphyseal dimensions (FAPDd, FTDd, FTDp; for statistical procedures M ATERIAL AND METHODS see Moncunill-Solé et al. 2018). In the case of extant Ochotona individuals, the weight was obtained from specimens’ tags and Material, measurements, and reconstruction of body mass museum records. It is well known that BM is strongly correlated and longevity with several life-history traits (Calder 1984, Peters 1986), so the The femur is considered the most suitable bone for carrying out longevity of the assessed P. sardus specimens was inferred using skeletochronology techniques in small mammals (Barker et al. their BMs and the allometric regression models proposed by 2003, García-Martínez et al. 2011, Kolb et al. 2015, Moncunill- Blueweiss et al. (1978). These allometric predictions (Table 1) Solé et al. 2016). In this regard, the present study involved 15 were compared with the longevity values gained from the histo- femora of Prolagus sardus of different dimensions and plate fu- logical study, providing a scaling context. sion status (Table 1; Supporting Information, Table S1), that formed an ontogenetic series ranging from juveniles to adults. Preparation and imaging of histological slides These fossil bones were retrieved from the Late Pleistocene Histological thin sections were obtained following standard- fossiliferous red clay layer of the Grotta della Medusa (NW ized protocols (García-Martínez et al. 2011, Kolb et al. 2015, Sardinia, Italy), also known as Dragonara Cave in the palae- Moncunill-Solé et al. 2016, Orlandi-Oliveras et al. 2016, ontological literature (Palombo and Rozzi 2014, Palombo et al. Miszkiewicz et al. 2020, Miszkiewicz and van der Geer 2022). 2017). Prolagus sardus material is curated in the ‘Laboratorio Each bone was completely embedded in epoxy resin (EPOFIX di Paleontologia dei Vertebrati, Università Rome Tre’ (LPV, Struers) due to its small size. For the correct interpretation of Life history of Prolagus sardus • 3 Table 1. Biological (age category, body mass, and allometrical longevity) and histological details (number of LAGs) relative to the assessed femora of Prolagus sardus (N = 15), Oc. collaris (N = 5), Oc. dauurica (N = 5), and Oc. princeps (N = 3). Specimens with bone tissues damaged by microorganisms are underlined. Abbreviations: no = no observable/not present, see others in the main text. Species Specimen Age BM (in g) LAGs Allometrical longevity (in years) P. sardus R129 J 0 P. sardus R77 J 0 Downloaded from https://academic.oup.com/zoolinnean/advance-article/doi/10.1093/zoolinnean/zlad112/7281581 by guest on 27 September 2023 P. sardus R113 J 0 P. sardus R137 J no P. sardus GD52 J 0 P. sardus R000 J 1 P. sardus R136 J/Y 2 P. sardus R30 J 2 P. sardus GD19 J no P. sardus D19 J no P. sardus R12 J/Y 2 P. sardus A17 Y 662.27 3 5.21 P. sardus R44 Y 569.11 2 5.07 P. sardus R5 A 663.09 8 5.21 P. sardus B045 A 730.28 no 5.29 Oc. collaris UAM 2929 A 124.10 2 Oc. collaris UAM 2947 A 126.90 1-2? Oc. collaris UAM 58315 Y 67.00 0 Oc. collaris UAM 63933 A 121.20 0-1? Oc. collaris UAM 63937 A 122.30 1 Oc. dauurica MSB 215075 Y 81.00 0 Oc. dauurica MSB 215674 J 48.00 0 Oc. dauurica MSB 215680 Y 139.00 0 Oc. dauurica MSB 215940 J 25.00 0 Oc. dauurica MSB 215953 A 141.00 0 Oc. princeps UAM 35060 A 130.00 0-2? Oc. princeps UAM 35119 A - 1 Oc. princeps UAM 13936 A 144.00 4 the section, the orientation of the bone was recorded through transmitted light microscope (Nikon Eclipse E600, combined the whole process by marking mediolateral and antero-posterior with a digital camera ZEISS AxioCam ICc3) and polarized light axes. Once embedded, the femoral midshaft (the surface of ones with lambda filters (Optika B-500POL and Leica Leitz interest) was exposed using a Struers’ Accutom-2 precision saw DMR, with digital camera Nikon DS-Fi1). Slides were photo- cutter. The obtained resin block was polished on a glass sheet graphed using the 10× lens for full image composition, and re- coated with carborundum powder with decreasing particle size gions of interest were captured using 20× and 40× lenses. (from 600 up to 1000 grit) and then fixed to a glass slide using the epoxy adhesive EPOTEK 301. The thin histological sections Osteohistology and growth marks were obtained with the precision saw cutter (0.5 mm cutting Bone tissues were typologically classified following the cri- disk) to a final thickness of about 100–110 μm, and were pol- teria of de Ricqlès et al. (1991), de Margerie et al. (2002), and ished again with a gradient of carborundum (800–1200 grit) by de Buffrénil et al. (2021a), considering mainly bone matrix and hand. Finally, they were dehydrated through a graded series of vascularization. Primary bone tissues can be remodelled by sec- alcohol baths, cleared in Histo-Clear II for 5 min, and covered ondary osteons (= Haversian bone), surrounded by cement lines with coverslips (100 μm) fixed with UV light-cured acrylic ad- (Padian and Lamm 2013). Besides, growth marks can be iden- hesive (Loctite 350). Histological slides of P. sardus were carried tified in bone tissues, which record changes in the bone depos- out at the Key Laboratory of Vertebrate Evolution and Human ition rate and allow age-related life-history traits to be inferred Origins (Chinese Academy of Sciences, China), whereas those (Castanet et al. 1993). Of particular interest are cyclical lines of of extant Ochotona at the ‘Servicio de preparación de rocas y arrested growth (LAGs) as they record annual growth cycles materiales duros, impresión y escaneado en 3D’ (Universidad de in lagomorphs (Millar and Zwickel 1972, Ohtaishi et al. 1976, Zaragoza, Spain). All of them are stored at the ‘Departamento de Frylestam and von Schantz 1977, Henderon and Bowen 1979, Física e Ciencias da Terra, Universidade da Coruña’ (UDC, A Iason 1988). These cyclical bone growth marks are ever-present Coruña, Spain). Histological thin sections were observed under in mammals and are related to photoperiod, metabolic rates, and 4 • Fernández-Bejarano et al. hormone cues (Castanet et al. 2004, Köhler et al. 2012). In this MA (Skedros et al. 2003, Miszkiewicz and van der Geer 2022), regard, the individual’s age at death (= longevity) can be inferred using the Analyze Particle tool of ImageJ 1.53k software. by counting the number of cyclical LAGs throughout the whole Descriptive statistical analyses of the variables were carried primary bone (Castanet et al. 2004, Castanet 2006, Kolb et al. out, testing the normality (Shapiro–Wilk test and Q–Q plots) 2015, Moncunill-Solé et al. 2016). Concerns associated with and equality of the variances (Levene’s test). The most appro- this skeletochronological approach, such as secondary bone or priate test of equality of means (T-test, ANOVA, Kruskal, Wallis, medullary resorption, are considered of lesser importance in etc.) was used to identify differences in CA, MA, TA, DAPm, Downloaded from https://academic.oup.com/zoolinnean/advance-article/doi/10.1093/zoolinnean/zlad112/7281581 by guest on 27 September 2023 lagomorphs, mainly as a consequence of their relatively short life and DTm: (i) among age categories ( J, Y, and A) of P. sardus; and and small size. On the other hand, non-cyclical marks can also (ii) among species (only adults of Ochotona spp. and P. sardus). appear in the bone cortex, reflecting ontogenetic events related Due to the noticeable size differences among groups, we also to biological stress (e.g. neonatal or weaning lines), remodel- analysed the ratios CA/TA and CA/MA, which remove the ef- ling processes (e.g. resorption lines), and fibres’ anchorage (e.g. fect of body size. In addition, ANCOVA tests were conducted Sharpey’s fibres), among others (Castanet et al. 2004, Nacarino- to assess statistically the existence of differences in the growth Meneses and Köhler 2018, de Buffrénil and Quilhac 2021). A trajectories among species (dependent variables: CA, MA, TA; neonatal line marks the beginning of postnatal cortical growth. covariables: DAPm and DTm, proxies of body size; and factor In mammals, an abrupt transition in the type of bone tissue is or independent variable: group). Prior to the analyses, the data observed following birth, documenting acceleration in growth were logarithmically transformed. The level of significance of all rate (Calderón et al. 2021). At present, no neonatal lines have statistical tests was fixed at 0.05 and was conducted using the been reported in fossil bones, though they are common in extant software jamovi v.2.3.24.0 (The jamovi Project 2022). squamates and have been identified in other tetrapods, including mammals (Nacarino-Meneses and Köhler 2018, de Buffrénil et RESULTS al. 2021b). A weaning line occurs as a consequence of dietary shift stress, and in large mammals it separates the pre-weaning Osteohistology and skeletochronology tissue, with a high growth rate, from post-weaning tissue, which Four of the P. sardus femora were attacked by microorgan- is deposited more slowly. This latter line is strongly marked in re- isms (e.g. bacteria, fungi, parasites, etc.) (Table 1; Supporting gions where the radial bone growth is constrained by the shape Information, Appendix S1), preventing their typological inter- of the bone, thereby giving rise to partial lines (Calderón et al. pretation and reducing the sample size to an ontogenetic series 2021). Besides, growth arrest is more pronounced during the of 11 individuals ensuring representation of all age classes. weaning stage (lasting 1–2 weeks) compared to birth (3 days), as observed in studies on other mammalian species, such as red Prolagus sardus from the Medusa Cave (Late Pleistocene, Sardinia, deer (Calderón et al. 2021). Italy) The following abbreviations were used to describe histological The youngest specimens (unfused epiphyses; R129, R77, and thin sections: cement line, CL; external fundamental system, R113) show an early ontogenetic stage (Fig. 1A, B), charac- EFS; fibrolamellar bone complex, FLC; Haversian system; HS; terized by a fast-growing bone formed mainly by FLC without inner cortical layer, ICL; lamellar bone, LB; lines of arrested LAGs. Vascularization, consisting of SV and PO, is abundant growth, LAGs; medullary cavity, MC; neonatal line, NL; non- and mostly oriented longitudinally (parallel to the proximodistal cyclical growth mark, nonCGM; non-vascular bone tissue, NV; axis). This early stage of rapid growth rate is especially remark- parallel-fibred bone, PFB; primary osteon, PO; resorption line, able in the smallest specimen (R129; FLmax = 28 mm), in RL; Sharpey’s fibres, SF; secondary osteon, SO; simple vascular which slow-growing bone traces have not been recognized (Fig. canal, SV; weaning line, WL; woven-fibred bone, WB. 1A, B). In R129 it is also possible to identify a partial nonCGM in the posterior region of the bone cortex, which we have related Quantitative geometrical analysis to an event of biological stress (likely a NL or WL) due to its Composition of the femur cross-sections were carried out main traits (position, sharpness, etc.) and resemblance to pre- by aligning and merging several images (10×) using Adobe vious reports (Fig. 1B, white arrowhead). Photoshop CS6 v.13.0 × 64 (1990-2012 Adobe Systems Later in ontogeny, but still during active growth (GD52, Incorporated). The images were converted into greyscale, 8-bit R000, and R30, with unfused epiphyses), an early periosteal de- format. For calculating cross-sectional geometrical data, we position of poorly vascularized PFB with longitudinally oriented used ImageJ 1.53k and the standardized protocol described by SV is observed (Fig. 1C). PFB marks a decrease in the growth Miszkiewicz and van der Geer (2022). Following the scale set- rate and a higher degree of tissue organization. Around the ting, the margins were cropped, positioning the femur cross- same time as PFB appears, the FLC is known to begin its re- sections in the centre. The noise points (e.g. resin bubbles), in sorption in the inner region of the cortex, expanding the MC, or outside the medullary cavity, were eliminated using the free- and depositing there an intermittent ICL of endosteal LB (Fig. hand selection tool. This selection tool was also used to trace the 1D). In the next stage, LB is deposited in the outer cortex of cortical area, and pixels were painted using black colour. Finally, the posterior region, before the attainment of skeletal maturity. the image was binarized using the threshold option, which forces This condition is illustrated in Fig. 1D (depicting the posterior greyscale pixels to black or white ones. On the transformed im- region of R000, a relatively large-sized juvenile, with unfused ages of the cross-sections of each specimen, we gained the fol- epiphyses), where the cortical region of FLC surrounded by a lowing geometrical data: cortical area (CA), medullary area very thin LB can be observed. At this moment, LAGs begin to (MA), and total area (TA); and the indexes CA/TA and CA/ appear in the outermost cortex (Fig. 1E, F), while the growth Life history of Prolagus sardus • 5 Downloaded from https://academic.oup.com/zoolinnean/advance-article/doi/10.1093/zoolinnean/zlad112/7281581 by guest on 27 September 2023 Figure 1. Bone histology of P. sardus specimens. A, B, R129 (juvenile, 0 LAG) showing the anterior region (A) formed by FLC with SVs and POs, and posterior one (B) where a nonCGM was identified (arrowhead). C, GD52 (juvenile, 0 LAG), medial region showing early external deposition of PFB with some SVs. In the inner cortex, WB is visible, as well as FLC with POs and SVs. D, R000 (juvenile, one LAG), posterior region with FLC sandwiched between the ICL and the outer cortex of LB (reversed image). E, R136 (young adult) showing two LAGs (arrowheads). F, A17 (young adult) with three LAGs (arrowheads). G, R30 (juvenile, two LAGs), detail of the lateral region with SOs 6 • Fernández-Bejarano et al. rate should sensibly decrease around its deposition. In the largest young adults, and adults considering the fusion status of the juveniles suitable for the analyses (R000 and R30, with unfused growth plates (Table 1; Supporting Information, Appendix S1). epiphyses; since those that actually were the largest—GD19 and The vast majority of the bones of juveniles and young adults D19—could not be analysed due to tissue damage), one to two are formed by FLC/WB with a high number of osteocytes and LAGs are visible, evidencing late epiphyseal closure compared vascular canals (Fig. 2A, B). The external periosteal layer is per- to extant species (Table 1). In this age category, we observed the vaded with abundant SVs, mainly longitudinally and radially highest histological variability among individuals. For example, oriented (Fig. 2B). None of the femora showed LAGs or growth Downloaded from https://academic.oup.com/zoolinnean/advance-article/doi/10.1093/zoolinnean/zlad112/7281581 by guest on 27 September 2023 although R000 and R30 have unfused epiphyses and similar size marks, except RLs. This suggests that all the specimens died be- (FLmax c. 40–43 mm), they present differences at the histo- fore reaching the first year of life. Despite this, we can note an logical level (e.g. R30 is strongly affected by remodelling, with ICL of slow-growing bone and PFB in some cortical regions of clear RL, etc.). The strong muscular anchorages, noted in the the adult specimen (MSB 215953, 0 LAG), as well as scarce and posterior region, increase the intragroup histological variation. punctual outer deposit of LB in the lateral region (Fig. 2C). These appear along with large SOs in the FLC, pointing out re- On the other hand, Oc. collaris (N = 5) and Oc. princeps modelling events (Fig. 1G). (N = 3) show a very distinct histological typology. All of these In the young adults (A17 and R44, showing fused epiphyses individuals are adults, except a specimen with suture lines with suture lines), HSs are recognized in several regions of the (UAM 58315, Oc. collaris). Poorly vascularized PFB is the most cortex. Particularly, they dominate the lateral area, whereas in common tissue observed in the histological slides—deposited the posterolateral one an external layer of highly organized PFB in the external and internal cortices—besides a middle layer of appears (Fig. 1H). A clear RL is identified in the outer cortex, FLC, sandwiched between them (Fig. 2D). The vascularization separating the PFB or EFS from the rest of bone tissues. These of this FLC is particularly abundant in some specimens (UAM specimens presented two to three LAGs, pointing to an age at 35060, Oc. princeps), with longitudinal and radial SVs (Fig. 2E). death of c. 2–3 years (Fig. 1F; Table 1). In the largest young adults The nearly adult specimen (UAM 58315, Oc. collaris), without (A17, with a minimal count of three LAGs) and in adults (R5; LAGs, shows a much reduced region of external PFB, signalling see next paragraph), the EFS is present (Fig. 1I). Accordingly, the earliest deposition of slow-growing bone before the epi- based on the number of observed LAGs and the moment of ap- physeal plate fusion. Most of the other specimens have at least pearance of the EFS, maturity is probably attained at c. 3 years one LAG (Fig. 2D), with a maximum of four recorded in UAM old. Two broken specimens (R136 and R12) have an inter- 113936 (Oc. princeps) (Fig. 2F). This oldest specimen also shows mediate status between juveniles and young adults when con- an ICL in the anteromedial region. The absence of LAGs in some sidering growth plate fusion and size. At the histological level, adult femora (fully fused epiphyses) points to an attainment of they have also an intermediate typology, shifted towards nearing skeletal maturity before the first year of life. In most of the histo- maturity (higher LB presence and >1 LAG) (Fig. 1L; Table 1). logical sections, we could observe RLs separating the PFB from In the adults (R5, with fused epiphyses), PFB and LB are the the internal layer of FLC. The EFS is not visible in any of them, most common tissues, occurring in most of the outer and inner LB is absent or very rare in all the specimens. cortices in the posterolateral and anteromedial regions, respect- ively (Fig. 1J). A wider EFS is noted externally, and PFB surrounds Quantitative geometrical analysis the SOs situated in the innermost cortex. There is a higher appos- Geometrical and size data were successfully gained from the as- itional rate of the EFS and ICL at the anterior region, with higher sessed specimens (see raw data in Supporting Information, Table rates of remodelling (Fig. 1J). In the examined sample, a max- S2), albeit BM reconstruction of P. sardus was only possible in imum of eight LAGs are identified (Fig. 1K; Table 1), suggesting a unbroken fused (or nearly fused) femora, belonging to young minimum age of 8 years at death. Such a specimen (R5) shows an adults or adults. The CA values range from 3.39 to 8.86 mm2 in important vascularized growth band between the fourth and fifth P. sardus, and from 0.54 to 2.43 mm2 in Ochotona (Fig. 3A, D), LAGs, which we hypothesize to be related to a particular mus- noting a lesser variation in the MA results (from 3.30 to 7.53 mm2 cular traction in the region. Accordingly, we can also observe SF in P. sardus individuals, and 1.41 to 2.13 mm2 in Ochotona). The through the whole of the outer cortex (Fig. 1K), with the highest extant and extinct adult specimens show, on average, a CA/MA presence in the medial and lateral regions. The muscular attach- index of 1.11 and 1.32, respectively, and CA/TA index of 0.52 ment is especially remarkable in the medial side, coinciding with and 0.57 (Fig. 3B, E). When assessing size variables, the values of the internal growth band of the posteromedial region. DTm are slightly larger and are more clustered around the mean than those of DAPm (Fig. 3C, F). In general, the distribution Extant relatives: Ochotona collaris, Ochotona dauurica, and of the log-transformed variables do not deviate from normality Ochotona princeps (P > 0.05; with the exception of logCA/MA and logCA/TA in The five specimens of Oc. dauurica show mostly fast-growing Oc. collaris adults), though it cannot be calculated in groups of bone tissues, though among them we can recognize juveniles, very small sample size (e.g. adult P. sardus, N = 2) (Supporting and SFs. H, R44 (young adult, two LAGs), lateral region with SOs and SFs, what prevents to determine the primary bone tissue. A clear RL was also noted. I, A17 (young adult, three LAGs), posteromedial region displaying a clear EFS formed by compact LB. J, K, R5 (adult, eight LAGs) showing the anteromedial region ( J) with FLC surrounded by LB in the inner and outer cortex and posteromedial region (K) with eight LAGs (arrowheads). L, R12 (young adult, with broken epiphyses) showing two LAGs (arrowheads). For abbreviations, see the text. Scale bars equals 100 μm. Life history of Prolagus sardus • 7 MC A FLC B C MC ICL FLC FLC LB Downloaded from https://academic.oup.com/zoolinnean/advance-article/doi/10.1093/zoolinnean/zlad112/7281581 by guest on 27 September 2023 WB MC SVs D E PFB F PFB FLC FLC SO PFB PFB ICL FLC WB ICL MC Figure 2. Bone histology of Ochotona specimens. A–C, Oc. dauurica. A, MSB 215940 (juvenile) showing a cortex formed by FLC and WB. B, MSB 215680 (young adult), with abundant SVs in the outer cortex. Note that microorganisms attacked this region, hiding bone tissues. C, MSB 215953 (adult) with FLC sandwiched between ICL and a scarce LB layer. Note the strong RL (black arrowhead). D, Oc. collaris UAM 63937 (adult), with an extensive deposition of PFB and clear RL (black arrowhead) splitting it from FLC. Notice the presence of one LAG (white arrowhead). E, F, Oc. princeps. E, UAM 35060 (adult), anterior region with PFB surrounded by a FLC full of SOs. F, UAM 113936 (adult), with detail of the PFB region, showing four LAGs (white arrowheads). For abbreviations, see the text. Scale bars equals 100 μm. Information, Appendix S1). Hence, the subsequent statistical S1). Overall, the statistical study reveals heterogeneity of the tests for these variables and groups are non-parametric. slopes in most of the assessed cases [interaction of the fix factor Statistical tests signal significant differences among age and covariable (the size), P < 0.05; Supporting Information, categories of P. sardus considering MA, TA, and DTm (P < 0.05; Appendix S1]. That is, the growth trajectories (relation be- Supporting Information, Appendix S1). Particularly, juveniles tween the dependent variable and the covariable) of these two show smaller values than the other groups (Fig. 3A, C), whereas groups are different (Fig. 3H, I). In only two cases can we as- the results of young adults and adults are very close. These dif- sume homogeneity of slopes (interaction P > 0.05), and only in ferences are related to a size effect, since the ratios CA/TA and one of them does the ANCOVA test fail to identify significant CA/MA are indistinguishable among age categories (P > 0.05; difference among groups (Fig. 3G). In this regard, results show Fig. 3B). When adults of extant ochotonids and P. sardus are that P. sardus has larger values of MA and TA than an average ex- considered, differences are observed in MA and TA (P < 0.05; tant Ochotona of similar size (Fig. 3H, I). In the case of CA, the Supporting Information, Appendix S1). In this case, Oc. collaris results are more doubtful (being similar in size between groups shows smaller values than P. sardus (Fig. 3A, D). However, con- or a slightly smaller in the case of P. sardus) (Fig. 3G; Supporting sidering that there are not statistical differences among adults of Information, Appendix S1). the three extant species (excepting TA, P ≈ 0.038; Fig. 3D–F), we prefer to include all of them in a sole group (Ochotona) for DISCUSSION running statistical tests with groups of larger sample size. In this latter case, results reveal that extant ochotonids and P. sardus Timing of Prolagus sardus life-history events: osteohistology have significant differences in all geometrical and size variables and skeletochronology (P < 0.05; Supporting Information, Appendix S1). However, The evaluation of the P. sardus juvenile specimens revealed that the absence of differences in the ratios CA/MA and CA/TA in- during early ontogeny the main bone tissue is a highly vascular- dicate that they are probably the result of a size effect between ized FLC, in line with previous investigations (Kolb et al. 2015, groups (Fig. 3B, E). Moncunill-Solé et al. 2016). At this stage, a nonCGM related to The growth trajectories are drawn using the dependent vari- a stress period was recognized in the smallest individual. These ables CA, MA, and TA, and the groups (fix factor) P. sardus and kinds of marks, described as incomplete or partial circumfer- Ochotona (all species in a sole group, see previous paragraph). In ences (Calderón et al. 2021), have been previously observed and all of the cases, the ANCOVA model is significant (P < 0.001), recorded only in very few extant mammalian taxa, highlighting as well as size proxy (DAPm or DTm) denoting a relationship their singularity (Morris 1970, Castanet et al. 2004, Nacarino- with fix factor (P < 0.01; Supporting Information, Appendix Meneses and Köhler 2018). In our case, we did not evidence 8 • Fernández-Bejarano et al. Downloaded from https://academic.oup.com/zoolinnean/advance-article/doi/10.1093/zoolinnean/zlad112/7281581 by guest on 27 September 2023 Figure 3. A–F, boxplots of log-transformed geometrical (CA, MA, CA/MA, and CA/TA) and size variables (DAPm and DTm). A–C, Prolagus sardus age categories ( J, Y, and A). D–F, adults of Oc. princeps, Oc. collaris, and Oc. dauurica. See Supporting Information, Table S2 for the raw data, including mean and standard deviation for species and age category. G–I, growth trajectories of CA, MA, and TA, considering DTm (size proxy), of P. sardus (N = 15) and Ochotona (N = 13). See Supporting Information, Appendix S1 for statistical results. major differences in bone tissue type following the deposition (Allen, 1890), respectively ( Jones et al. 2009). Large ratios (= of this nonCGM, ruling out its attribution to a NL (Calderón et larger, more fully developed pups) are related to more precocial al. 2021). By contrast, its characteristics and resemblance to pre- offspring (large size at birth, small litters, or monotocous, etc.) vious evidence allow us to tentatively relate it to a WL (Castanet (Dobson and Oli 2007), which in general suggest a slow pace of et al. 2004). Considering the diameter of this nonCGM (≈2.4 life (Gaillard et al. 1989, Bielby et al. 2007, Trillmich et al. 2019). mm), and the midshaft diameters of adults, the weaning should The presence of a WL is quite exceptional evidence, never de- occur when the juvenile achieved 55–60% adult size (≈375–425 scribed in bones belonging to an extinct small mammal (<2 kg) g). This value is in agreement with those documented in Oc. so far. princeps (adult BM = 157.63 g and BM at weaning = 88.49 g; Later in ontogeny, different steps towards a reduction of the 56%) but are far from the 79% observed in Lepus alleni Mearns, growth rate were identified in P. sardus. First, in middle-sized ju- 1890 and L. californicus Gray, 1837, and from the 13% and 9% of veniles, the tissues become more organized with the deposition Oryctolagus cuniculus (Linnaeus, 1758) and Sylvilagus floridanus of PFB at endosteal and periosteal levels. Later, LB is deposited Life history of Prolagus sardus • 9 forming an intermittent ICL before the occurrence of the first In addition, it is greater than the largest P. sardus minimum age LAG. However, it is not until the approximately first year of life reported in previous studies (Kolb et al. 2015), and it is the that the first LB is deposited externally and the presence of EFS longest longevity documented in a species of Prolagus (consid- is only seen in young adults and adults. The decrease in growth ering mainland and insular populations) up to now (Kolb et al. rate, marked by deposition of slow-growing tissues, has been re- 2015, Moncunill-Solé et al. 2016). lated to maturity, though there is a lack of consensus about how this is recorded in bone microstructure. Generally, age at ma- Histology and life-history strategies of Ochotona Downloaded from https://academic.oup.com/zoolinnean/advance-article/doi/10.1093/zoolinnean/zlad112/7281581 by guest on 27 September 2023 turity, described as the periosteal apposition fall, is typically rec- Two major groups of extant pikas can be distinguished when ognized by EFS presence (Castanet et al. 2004, García-Martínez considering life-history patterns: (i) the talus group, and (ii) the et al. 2011). Whilst some authors have linked its deposition to meadow or steppe group, though they are all remarkably similar skeletal maturity (epiphyseal fusion) (Chinsamy-Turan 2005, at the morphological level (Smith 1988, Smith et al. 1990, 2018). Klein et al. 2012, Nacarino-Meneses et al. 2016, Woodward et al. Talus pikas are non-burrowing leapers that inhabit rock piles, 2020 and references therein), others also attributed it to sexual characterized by low extrinsic mortality (average mortality) and maturity as well (Köhler and Moyà-Solà 2009, Köhler et al. 2012, interspecific competition levels in which birth and death rates Marín-Moratalla et al. 2013). However, this last inference is not are in equilibrium (stable low population densities). Under confirmed as both maturities can be asynchronous (e.g. sexual this scenario, selection favours traits that maximize efficiency preceding skeletal one) (Martínez-Maza et al. 2014, Jordana et (K-selection strategy) (Pianka 2011). Hence, talus pikas show al. 2016). In small mammals, maturity (skeletal and sexual) is a low fecundity rates (two litters per season of few pups, and only a very difficult trait to trace, since, in general, their bones are sub- few juveniles are successfully weaned), later maturation (first re- jected to rapid growth rates and most life-history events occur production occurs not before reaching the first year of life), and a within the first year of life (Castanet et al. 2004, García-Martínez longer lifespan (Smith 1988). In burrowing pikas, opposite life- et al. 2011). history traits are observed (short-lived, fluctuating population Considering the aforementioned, our results suggest that P. densities, fast maturation, large juvenile production, etc.). They sardus’ skeletal maturity was attained at 3 years, though the first growth in habitats characterized by strong environmental per- slow-growing tissue appeared at the age of 1 year. Juveniles and turbations (e.g. flooding or cold temperatures) with high mor- most of the young adults (≤2 years) were in somatic growth (un- tality rates, such as meadows, steppes, and forests (Smith 1988). fused epiphyses) and EFS was not present. In the case of pikas, In these environments, the fitness of individuals is increased by there are no records about the pace and time of skeletal growth traits that maximize productivity (r-selection strategy) (Pianka plate fusion. Our Ochotona specimens with LAGs show fully 2011). In view of the above, talus pikas show a slower life history fused epiphyses, pointing to an earlier ending of somatic growth than the meadow ones. (i.e. earlier than the first year of life) with respect to P. sardus, In the present study, we have assessed the bone tissue of spe- regardless of their ecotype. The sexual maturity of these species, cimens from the two groups. Unfortunately, most of the life- however, is different: Oc. princeps and Oc. collaris (talus-ecotype) history traits detailed above are not recorded in their bones (e.g. attained it at 1 year after birth, and Oc. dauurica (meadow- number of pups per litter), but we were able to identify some sig- ecotype) at 4–5 weeks (Smith and Weston 1990, Matsuzaki et al. nificant differences. In the case of Oc. dauurica (meadow group), 1998, Lanier and Hik 2016). In the case of P. sardus, we do not the absence of LAGs and the fusion of epiphyses point to an have enough evidence to determine whether skeletal maturity early attainment of skeletal maturity (before the first year of life), and sexual maturity were synchronous or not. Recent studies agreeing with an r-strategy. Based on the literature, Oc. dauurica on the age of maturity in the giant insular leporid Nuralagus rex females reach sexual maturity very rapidly, being able to produce Quintana et al. (2011) (Pliocene of Menorca, Spain; BM = 8 a litter in the same year of their birth. Their pups are altricial kg, medium mammal) also signal a delayed maturation on the (naked, without fur, closed eyes, etc.) and weaned at 1 month basis of EFS presence, though they correlate it with sexual mat- old (Sokolov et al. 1994, Ciwko 2014). In the wild, they do not uration (Köhler et al. 2022). Based on this, Riyahi et al. (2011) survive more than 1 or 2 years (representing the 14% and 3% of determined a sexual maturity of this species at 2–3 years of age, the population, respectively) (Smith et al. 2018). The absence of whereas Köhler et al. (2022) suggested that females attained a large amount of slow-growing tissue in their bones also points it at age of 4 years. If we consider leporids, available data indi- to a fast growth rate. Conversely, the deposition of slow-growing cate that the distal femoral epiphyseal closure of Or. cuniculus tissue is common (even this appears in the youngest individual (breed New Zealand; BM = 4.5–5.4 kg) occurs at an average with suture lines and remodelling fast-growing one) in the bones age of 5.25–5.50 months (radiographically and histologically) of Oc. collaris and Oc. princeps (talus group), but the skeletal ma- (Kaweblum et al. 1994). In the case of S. floridanus (BM = 1.2 turity is attained also before the first year. However, they are kg), long bone epiphyses fusion occurs at age of ≈ 4.5 months longer lived specimens (one to four LAGs in adults). According (133 days) (Young et al. 2022). Thus, we can affirm that skel- to our results, Oc. collaris reaches the adult size very rapidly (40– etal maturity was attained quite slowly in P. sardus considering its 50 days) and is sexually reproductive as yearlings. This species BM (c. 700–750 g) (Moncunill-Solé et al. 2021). has a relatively long lifespan (4–7 years) (Leininger 2009, Smith Some P. sardus young adults and adults have important regions et al. 2018). In the case of Oc. princeps, these events occur simi- of LB and SOs (Haversian systems) in the innermost cortex. larly. Sexual maturity of males and females is documented on The largest specimen had eight LAGs in the EFS, pointing to a average at 1 year (in the summer after birth), though they reach minimum age at death of 8 years. This minimum longevity is much the adult size in only 3 months, and their lifespan ranges from 3 extended than the 5 years estimated by its adult BM (Table 1). to 8 years (Peri 2012). 10 • Fernández-Bejarano et al. When comparing bone tissues of P. sardus and long-lived the digging specialization in small mammals failed to find a rela- Ochotona, clear differences concerning the grade of vasculariza- tionship with compactness (Meier et al. 2013), though thicker tion and tissue type were observed. In pikas of slow life history bone walls have been documented (Montoya-Sanhueza and (excepting the youngest specimens), vascularization is scarce, Chinsamy 2016). The lifestyles of P. sardus and pikas (particu- whereas in all age categories of P. sardus, SVs and POs are very larly those from the talus) are not very different. Prolagus sardus abundant. This distinction could be related to the higher meta- was not a good runner, but had skills for digging, climbing, and bolic needs of larger species. Avascular bones are common in scrambling (Dawson 1969, Moncunill-Solé 2021), and talus Downloaded from https://academic.oup.com/zoolinnean/advance-article/doi/10.1093/zoolinnean/zlad112/7281581 by guest on 27 September 2023 skeletons of small vertebrates, because superficial capillaries are pikas show a general reduction of their cursoriality but are good enough to supply the vascular demands of tissues (de Buffrénil leapers (Reese et al. 2013, Young et al. 2014). Our preliminary et al. 2021c). The other main difference is the amount of slow- results do not clarify if the differences in cortical and medullar growing bone: LB hardly ever appears in thin sections of adult proportions between these two clades may be related to phyl- Ochotona but is frequent in those of P. sardus. This signals that a ogeny or other factors (small sample, environmental effect, etc.). slower growth rate probably occurred in P. sardus at adult stages (Amprino 1947, de Margerie et al. 2002, 2004). The results Insights into the life-history strategy of the ‘giant’ small in- and comparisons are preliminary, as they are based on a limited sular mammals sample of pikas not covering their ontogenetic variation. Future Insular environments are recognized ecologically by their re- studies involving samples with a complete ontogenetic range are duced resource levels and lower extrinsic mortality (Sondaar needed to describe in detail the variation of bone tissues in pikas 1977). In this insular ecological context, and following the and improve the comparisons with their relative P. sardus. theoretical models (based on Life History Theory; Palkovacs 2003), selection promotes a shift in life-history strategy, though Quantitative geometrical analysis the relative influence of these selection pressures depends on the Size (or its proxy, BM) covaries with an extensive set of mor- taxon size. In the case of small mammals, the low extrinsic mor- phological, behavioural, physiological, life-history, and eco- tality is the primary driver. Their fitness is increased by maturing logical variables, and when organisms change in size (scaling) later (increased age at maturity and longevity) and at a larger they have to face structural, functional, and biological conse- adult size, which promotes an increased fecundity (probably in quences (Calder 1984, Schmidt-Nielsen 1984, Peters 1986). the size of the offspring, being more competitive). That is, selec- For this reason, it is relevant to consider size in comparative tion favours a shift towards the slow end of the life-history con- studies (especially with marked size differences, as it is our tinuum, being more efficient in resource acquirement and more case) (Glazier 2022). Our results do not identify difference competitive, and maximizing lifetime reproduction (allocating among geometrical variables (CA, MA, TA) when size effect less energy to early reproduction). Empirical evidence reinforces is removed (ratios CA/MA and CA/TA), suggesting that pro- these predictions, both in extant and extinct biotas (Adler and portions of cortical and medullary cavity remain stable in all Levins 1994, Fons et al. 1997, Palkovacs 2003 and references groups (ontogenetic groups of P. sardus; and interspecific groups therein). Although this general life-history explanation accounts of adult Ochotona and P. sardus). However, significant differ- for the widely observed trends, there is no consensus with other ences were identified between growth trajectories (considering biological disciplines. In this regard, community ecologists work juveniles, young adults, and adults): P. sardus shows slightly on demonstrating that selection pressures (predation, competi- thinner cortex and a proportionally larger medullary cavity tion, and resource) operate directly on body size and other bio- than ochotonids. Thicker cortices in terrestrial amniotes have logical traits of insular dwellers (Heaney 1978, Lomolino 2005, been related to more resistance to bending and torsion forces/ Durst and Roth 2015; among others). loads, and cortex % (at mid-diaphysis of limb bones) has been The timing of P. sardus life-history events recorded in our correlated with body mass (weight bearing support), ecology, study points to a slow pace of life, which improves the fitness of locomotor activities (swimming, cursorial, etc.), or phylogeny the species, and agrees with the models proposed by Life History (Currey and Alexander 1985, Francillon-Vieillot et al. 1990, Ruff Theory and previous studies (Moncunill-Solé et al. 2021, in 2000, Cubo et al. 2005, Legendre et al. 2012, Quemeneur et al. press). Although we were not able to determine the mass at birth 2013, Canoville et al. 2022). Overall, terrestrial taxa are charac- or age at weaning, our results suggest that weaning occurred at terized by moderate cortical thickness with a well-differentiated large size (55–60% of adult size, ~375–425 g), which could point medullary cavity (without bony trabeculae), aquatic forms by to precocial pups. Accordingly, the Corsican population of the more compact, dense bones (appendicular long bones, ribs, etc., lesser white-toothed shrew produces, statistically, litters of fewer up to ~90%) without a free medullary cavity, and semi-aquatic pups but of larger size than mainland relatives (Fons et al. 1997). or amphibious species by intermediate traits (thicker cortices Larger offspring were also observed in the insular bank vole and more trabecular network, though overlapping with those of populations (Mappes et al. 2008), as well as in the insular Santa terrestrial relatives) (Laurin et al. 2011, Canoville et al. 2021). Catarina’s guinea pig (19% mass of adult female), concurrently By contrast, the cortices of land-dwelling graviportal taxa and with smaller litter sizes (Salvador and Fernandez 2008). Insular large terrestrial birds (extinct) are thicker and the medullary opossums (Austad 1993) and amphibians (Wang et al. 2009, Xu smaller than general tetrapods (to a lesser extent than aquatic et al. 2019) pursue this trend too. In this regard, and following taxa) (Houssaye et al. 2016, Canoville et al. 2022), whereas in present evidence, it is likely that P. sardus gave birth to a lower the case of flying amniotes, the cortical thickness is lower than number of pups per litter than what is expected allometrically, terrestrial ones and show expanded medullary cavities (Currey but of larger size. This improved its juvenile mortality rate and and Alexander 1985, Canoville et al. 2021). Researchers testing increased the fitness of the P. sardus population, which was near Life history of Prolagus sardus • 11 carrying capacity (higher population densities) (Stearns 1992). scientific community. The present paper contributes to this de- The somatic growth of the species was prolonged (3 years), bate assessing the life-history traits and strategy of the extinct in- though with present data we cannot affirm that the onset of re- sular ‘giant’ Prolagus sardus with a sample that spanned different production time was reached at such a late age. Allometrically, ontogenetic stages, retrieved from Grotta della Medusa (Late the species should mature sexually at 10 months (Blueweiss Pleistocene). The osteohistological and skeletochronological et al. 1978). The smallest Sylvilagus, similar in size to P. sardus, study of fossil femora revealed that: (i) P. sardus’ pups were reach sexual maturity between the fifth and sixth month of life weaned at large size; (ii) species’ maturation was delayed in its Downloaded from https://academic.oup.com/zoolinnean/advance-article/doi/10.1093/zoolinnean/zlad112/7281581 by guest on 27 September 2023 [S. bachmani (Waterhouse, 1839), 610 g, 154 days; S. audubonii life cycle; and (iii) P. sardus lived longer than expected for its (Baird, 1858), 756 g, 164 days; Tacutu et al. 2018]; whereas in size. Such biological traits are predicted in those insular species the case of long-lived pikas (130–200 g, from talus ecotype), more affected by the decreased mortality (mainly small-sized sexual maturity is reached as yearlings, contrasting with the 7 ones), and are typical of taxa that are more efficient in resource months expected allometrically (Oc. princeps, Moncunill-Solé et acquirement and allocated less energy to early reproduction al. 2016). Even though we cannot affirm the exact moment when (K-selection strategy). In this regard, it is probable that the pre- maturity occurred, it is likely that P. sardus age at maturity was dation pressure exerted by the Sardinian dhole was not so ex- also delayed to that expected allometrically. The later ontogen- treme as several researchers has proposed, and/or that P. sardus etic stage (adulthood) of the species was extended (minimum adopted several behavioural strategies to face them. On the other lifespan of 8 years), as occurred in other small and large insular hand, the comparative histological and quantitative geometrical mammals (Austad 1993, Köhler and Moyà-Solà 2009, Riyahi et analysis among Ochotona spp. and P. sardus identified clear dif- al. 2011, Kolb et al. 2015, Moncunill-Solé et al. 2016, Orlandi- ferences. Less vascularization and low presence of slow-growing Oliveras et al. 2016, Köhler et al. 2021 and references therein, bone tissue characterized the Ochotona femoral thin section, Miszkiewicz and van der Geer 2022). which have been tentatively related to its smaller size and faster Contrary to other insular small-sized mammals, P. sardus growth rate. In addition, they also differed in cortical and me- coexisted with some aerial and terrestrial predators in the Late dullary growth trajectories, which could be the consequence Pleistocene Sardinian ecosystems (birds of prey, small carni- of differences in body mass, ecology, locomotor activities, and vores, etc.), the most noteworthy of which is the Sardinian phylogeny. The results obtained in the present research concur dhole, Cynotherium sardous Studiati, 1857 (Palombo and Rozzi, with the eco-evolutionary responses described in extinct insular 2014). Although it was originally described as a specialized lagomorphs and other small mammals to date, and provide new Prolagus hunter (Malatesta 1970), the most recent studies have empirical evidence about evolution in insular domains not per- considered that the dhole preyed on different small-sized ani- turbed by human action. mals (including small mammals, birds, etc.) (Lyras and van der Geer 2006, Lyras et al. 2006). The life-history traits of P. sardus recorded in the present study (shifting towards the slow end) did SUPPLE MEN TARY DATA not follow the high predation risk mainland strategy and they are Supplementary data are available at Zoological Journal of the similar to those predicted when decreased mortality dominates Linnean Society online. (Palkovacs 2003). This suggests that the predation pressure of C. sardous towards P. sardus was not so extreme as some researchers ACKNOWLEDGE MEN TS have suggested and is lower than those recorded in mainland en- vironments. In fact, these studies did not show direct evidence The author thank Dr Jeffrey Streicher (Editor) and anonymous re- that P. sardus was the only target of this carnivore (Lyras and viewers for their constructive and valuables comments, which helped van der Geer 2006, Lyras et al. 2006). In addition, the species us improve this research. Besides, authors are indebted to Dr Link may have adopted several biological or behavioural strategies Olson and Aren M. Gunderson (Museum of the North Mammal to face predators (as occurs in talus pikas) (Schoenherr 1992). Collection, University of Alaska) and Dr Joseph Cook and Dr Jonathan L. Dunnum (Museum of South Western Biology, University of New The preliminary assessments of their bones, and those from Mexico) for providing the Ochotona material assessed in this paper; and their insular sister-taxa, suggest skills for digging and scrambling to Dr José Manuel Amado Paz for allowing access to the microscope rough terrains, where the individuals could hide (Dawson 1969, devices in Campus the Esteiro (UDC, Ferrol, Spain). In addition, au- Moncunill-Solé 2021). Such habitats (cliffs, high rocky coasts, thors would like to acknowledge the use of ‘Servicio General de Apoyo etc.) were common in Sardinia during the Last Glacial Maximum a la Investigación-SAI, Universidad de Zaragoza’ and Key Laboratory (age of the Grotta della Medusa), characterized by a low stand of Vertebrate Evolution and Human Origins, Chinese Academy of sea-level (c. 130 m depth below to the present-day level) and the Sciences. emergence of a great part of the present Mediterranean contin- ental shelf (Deiana et al. 2021). Future studies on postcranial FUNDING bones could draw out insights on the ecology and habitat pref- A.B. is supported by the ‘Ramón y Cajal’ fellowship (RYC2021- erence of P. sardus, as well as provide relevant data about the 034269-I), funded by MCIN/AEI/10.13039/501100011033 and the Sardinian ecological networks. European Union «NextGenerationEU»/PRTR; C.A. is supported by the Visiting Professor grant of the President’s International Fellowship Initiative of the Chinese Academy of Sciences, Spanish Agencia Estatal CONCLUSIONS de Investigación and the European Regional Development Fund of In recent years, the study of the eco-evolutionary adaptations of the European Union (CGL2016-76431-P), Cerca Program of the insular small mammals has generated a surge of interest from the Generalitat de Catalunya and the Grant to Department of Science, 12 • Fernández-Bejarano et al. Roma Tre University (CA, MIUR-Italy Dipartimenti di Eccellensa, Art. de Buffrénil V, Quilhac A, Castanet J. Cyclical growth and 1, C. 314-337 L. 232/016); and B.M.-S. by the Consellería de Cultura, skeletochronology. In: de Buffrénil V, de Ricqlès AJ, Zylberberg L, Educación e Universidade, Xunta de Galicia (ED481D-2022-013). Padian K (eds), Vertebrate Skeletal Histology and Paleohistology, 1st Funding for open access charge: Universidade da Coruña/CISUG. edn. Boca Raton: CRC Press, 2021b, 626–44. de Buffrénil V, Quilhac A, Cubo J. Accretion rate and histological fea- tures of bone. In: de Buffrénil V, de Ricqlès AJ, Zylberberg L, Padian CONFLICT OF INTEREST K (eds), Vertebrate Skeletal Histology and Paleohistology, 1st edn. Boca Raton: CRC Press, 2021c, 221–8. Calder WA. Size, Function and Life History. Cambridge: Harvard Downloaded from https://academic.oup.com/zoolinnean/advance-article/doi/10.1093/zoolinnean/zlad112/7281581 by guest on 27 September 2023 The authors declare that they have no known competing financial inter- ests or personal relationships that could have appeared to influence the University Press, 1984. work reported in this paper. Calderón T, Arnold W, Stalder G et al. Labelling experiments in red deer provide a general model for early bone growth dynamics in rumin- ants. Scientific Reports 2021;11:14074. DATA AVAIL ABILIT Y STATE MENT Canoville A, de Buffrénil V, Laurin M. Bone microanatomy and lifestyle in tetrapods. In: de Buffrénil V, de Ricqlès AJ, Zylberberg L, Padian The data that supports the findings of this study are available in the K (eds), Vertebrate Skeletal Histology and Paleohistology, 1st edn. Boca Supplementary Data of this paper. Raton: CRC Press, 2021, 724–43. Canoville A, Chinsamy A, Angst D. New comparative data on the long bone microstructure of large extant and extinct flightless birds. REFERENCES Diversity 2022;14:298. Adler GH, Levins R. The island syndrome in rodent populations. The Castanet J. Time recording in bone microstructures of endo- Quarterly Review of Biology 1994;69:473–90. thermic animals: functional relationships. Comptes Rendus Palevol Allen JA. Lepus alleni Mearns: antelope jack rabbit. Bulletin of the American 2006;5:629–36. Museum of Natural History 1890;2:294–7. Castanet J, Francillon-Vieillot H, Meunier F, De Ricqlés A. Bone and in- Alves PC, Hackländer K. Lagomorph species: geographical distribu- dividual aging. In: Hall BK (ed.), Bone: A Treatise, Vol. 7. Boca Raton: tion and conservation status. In: Alves PC, Ferrand N, Hackländer K CRC Press, 1993, 245–83. (eds), Lagomorph Biology: Evolution, Ecology, and Conservation. Berlin: Castanet J, Croci S, Aujard F et al. Lines of arrested growth in bone and Springer, 2008, 395–406. age estimation in a small primate: Microcebus murinus. Journal of Amprino R. La structure du tissu osseux envisagée comme expression Zoology 2004;263:31–9. de differences dans la vitesse de l’accroissement. Archives de Biologie Čermák S, Angelone C, Moncunill-Solé B. Prolagus Pomel, 1853 1947;58:315–30. (Lagomorpha, Mammalia) in the framework of the Pliocene Angelone C, Zhang ZQ. Climate change and evolution of early lago- faunal rearrangements in central Europe. Comptes Rendus Palevol morphs (Mammalia): a study perspective based on new materials of 2021;20:597–617. Ordolagus from Nei Mongol (northern China). Vertebrata PalAsiatica Chapman JA, Flux JEC. Rabbits, Hares and Pikas. Status Survey and 2021;59:138–68. Conservation Action Plan, 1st edn. Gland: International Union for Angelone C, Moncunill-Solé B, Kotsakis T. Contribution of fossil Conservation of Nature and Natural Resources, 1990. Lagomorpha (Mammalia) to the refinement of the late Miocene– Chinsamy-Turan A. The Microstructure of Dinosaur Bone: Deciphering Quaternary palaeobiogeographical setting of Italy. Comptes Rendus Biology with Fine-Scale Techniques. New Jersey: John Wiley & Sons, Palevol 2019;18:1025–40. 2005. Austad SN. Retarded senescence in an insular population of Virginia opos- Ciwko E. Ochotona dauurica, Animal Diversity Web. 2014. https:// sums (Didelphis virginiana). Journal of Zoology 1993;229:695–708. animaldiversity.org/accounts/Ochotona_dauurica/ (21 March Baird SF. Mammals. Reports of Explorations and Surveys, to Ascertain 2023, date last accessed). the Most Practicable and Economical Route for a Railroad from the Cubo J, Ponton F, Laurin M et al. Phylogenetic signal in bone microstruc- Mississippi River to the Pacific Ocean 1858;8:12–45. ture of sauropsids. Systematic Biology 2005;54:562–74. Barker JM, Boonstra R, Schulte-Hostedde AI. Age determination Currey JD, Alexander RM. The thickness of the walls of tubular bones. in yellow-pine chipmunks (Tamias amoenus): a comparison of Journal of Zoology 1985;206:453–68. eye lens masses and bone sections. Canadian Journal of Zoology Dawson MR. Osteology of Prolagus sardus, a Quaternary ochotonid 2003;81:1774–9. (Mammalia, Lagomorpha). Palaeovertebrata 1969;2:157–90. Bielby J, Mace GM, Bininda-Emonds ORP et al. The fast–slow continuum Deiana G, Lecca L, Melis RT et al. Submarine geomorphology of the in mammalian life history: an empirical re-evaluation. The American Southwestern Sardinian continental shelf (Mediterranean Sea): in- Naturalist 2007;169:748–57. sights into the Last Glacial Maximum sea-level changes and related Blueweiss L, Fox H, Kudzma V et al. Relationships between body size and environments. Water 2021;13:155. some life history parameters. Oecologia 1978;37:257–72. Dobson FS, Oli MK. Fast and slow life histories of rodents. In: Wolff JO, Bonino N, Cossíos D, Menegheti J. Dispersal of the Europoean hare, Sherman PW (eds), Rodent Societies. An Ecological and Evolutionary Lepus europaeus in South America. Folia Zoologica 2010;59:9–15. Perspective. Chicago: University of Chicago Press, 2007, 99–105. Bover P, Lira-Garrido L, Beltrán P, Alfaro-Ibáñez MP, Moncunill-Solé Durst PAP, Roth VL. Mainland size variation informs predictive models B, Angelone C. Mitochondrial phylogeny of the extinct Prolagus of exceptional insular body size change in rodents. Proceedings of the sardus (Lagomorpha, Mammalia): Reigniting the debate about Royal Society B 2015;282:20150239. family Prolagidae. In: Belvedere M, Mecozzi B, Amore O, Sardella Fenner F. Deliberate introduction of the European rabbit, Oryctolagus R (eds), Abstract Book of the XIX Annual Conference of the European cuniculus, into Australia. Revue Scientifique et Technique 2010;29:103–11. Association of Vertebrate Palaeontologists, Special Vol. 1. Benevento: Ferreira CC, Castro F, Piorno V et al. Biometrical analysis reveals major PalaeoVertebrata, 2022, 20–1. differences between the two subspecies of the European rabbit. de Buffrénil V, Quilhac A. Bone tissue types: a brief account of currently Biological Journal of the Linnean Society 2015;116:106–16. used categories. In: de Buffrénil V, de Ricqlès AJ, Zylberberg L, Padian Flynn LJ, Winkler AJ, Erbajeva M et al. The Leporid Datum: a late K (eds), Vertebrate Skeletal Histology and Paleohistology, 1st edn. Boca Miocene biotic marker. Mammal Review 2014;44:164–76. Raton: CRC Press, 2021, 147–82. Fons R, Poitevin F, Catalan J et al. Decrease in litter size in the shrew de Buffrénil V, de Ricqlès AJ, Zylberberg L, Padian K (eds.) Vertebrate Crocidura suaveolens (Mammalia, Insectivora) from Corsica (France): Skeletal Histology and Paleohistology, 1st edn. Boca Raton: CRC Press, Evolutionary response to insularity? Canadian Journal of Zoology 2021a. 1997;75:954–8. Life history of Prolagus sardus • 13 Francillon-Vieillot H, de Buffrénil V, Castanet J, Géraudie J, Meunier Kolb C, Scheyer TM, Veitschegger K et al. Mammalian bone FJ, Sire JY, Zylberberg L, de Ricqlès AJ. Microstructure and min- palaeohistology: a survey and new data with emphasis on island eralization of vertebrate skeletal tissues. In: Carter JG (ed.), Skeletal forms. PeerJ 2015;3:1358. Biomineralization: Patterns, Processes and Evolutionary Trends. Lanier H, Hik D. Ochotona collaris. The IUCN Red List of Threatened Washington: American Geophysical Union, 1990, 175–234. Species, 2016, eT41257A45182533. Frylestam B, von Schantz T. Age determination of European hares based Laurin M, Canoville A, Germain D. Bone microanatomy and lifestyle: a on periosteal growth lines. Mammal Review 1977;7:151–4. descriptive approach. Comptes Rendus Palevol 2011;10:381–402. Gaillard JM, Pontier D, Allainé D et al. An analysis of demographic tactics Leack K, Kelly R, Cameron A et al. Expertly validated models and in brids and mammals. Oikos 1989;56:59–76. phylogenetically-controlled analysis suggests responses to climate Downloaded from https://academic.oup.com/zoolinnean/advance-article/doi/10.1093/zoolinnean/zlad112/7281581 by guest on 27 September 2023 García-Martínez R, Marín-Moratalla N, Jordana X et al. The ontogeny of change are related to species traits in the Order Lagomorpha. PLoS bone growth in two species of dormice: reconstructing life history One 2015;10:e0122267. traits. Comptes Rendus Palevol 2011;10:489–98. Legendre L, LeRoy N, Martínez-Maza C et al. Phylogenetic signal in bone Glazier DS. Complications with body-size correction in comparative histology of amniotes revisited. Zoologica Scripta 2012;42:44–53. biology: possible solutions and an appeal for new approaches. Journal Leininger C. Ochotona collaris, Animal Diversity Web. 2009. https:// of Experimental Biology 2022;225:jeb243313. animaldiversity.org/accounts/Ochotona_collaris/ (21 March 2023, Gray JE. Description of some new or little known Mammalia, principally date last accessed). in the British Museum Collection. The Magazine of Natural History Linnaeus C. Systema naturæ per regna tria naturæ, secundum classes, ordines, and Journal of Zoology, Botany, Mineralogy, Geology, and Meteorology, genera, species, cum characteribus, differentiis, synonymis, locis; Editio New Series 1837;1:577–87. decima, reformata; Tomus I. Holmiæ: Laurentii Salvii, 1758. Gureev AA. Lagomorphs (Lagomorpha) from the Oligocene of Mongolia Lomolino MV. Body size evolution in insular vertebrates: generality of and Kazakhstan. Trudy Paleontologicheskogo Instituta Akademii Nauk the island rule. Journal of Biogeography 2005;32:1683–99. SSSR 1960;77:5–34 [in Russian]. López-Martínez N. The lagomorph fossil record and the origin of the Heaney LR. Island area and body size of insular mammals: evidence European rabbit. In: Alves PC, Ferrand N, Hackländer K (eds), from the tri-colored squirrel (Callosciurus prevosti) of southeast Asia. Lagomorph Biology. Evolution, Ecology and Conservation. Berlin: Evolution 1978;32:29–44. Springer, 2008, 27–46. Henderon BA, Bowen HM. A short note: estimating the age of the Lu D, Zhou CQ, Liao WB. Sexual size dimorphism lacking in small mam- European rabbit, Oryctolagus cuniculus, by counting the adhesion lines mals. North-Western Journal of Zoology 2014;10:53–9. in the periosteal zone of the lower mandible. Journal of Applied Ecology Lyras G, van der Geer A. Adaptations of the Pleistocene island canid 1979;16:393–6. Cynotherium sardous (Sardinia, Italy) for hunting small prey. Cranium Hoffmann RS, Smith AT. Order Lagomorpha. In: Wilson DE, Reeder 2006;23:51–60. DM (eds), Mammal Species of the World: A Taxonomic and Geographic Lyras GA, van der Geer AAE, Dermitzakis MD et al. Cynotherium sardous, Reference, 3rd edn. Baltimore, MA: Johns Hopkins University, 2005, an insular canid (Mammalia: Carnivora) from the Pleistocene of 185–211. Sardinia (Italy), and its origin. Journal of Vertebrate Paleontology Houssaye A, Waskow K, Hayashi S et al. Biomechanical evolution of solid 2006;26:735–46. bones in large mammals: a microanatomical investigation. Biological Malatesta A. Cynotherium sardous Studiati. An extinct canid from the Journal of the Linnean Society 2016;117:350–71. Pleistocene of Sardinia. Memorie dell’Istituto Italiano di Paleontologia Iason GR. Age determination of mountain hares (Lepus timidus): a rapid Umana Nuova Serie 1 1970;1:1–72. method and when to use it. Journal of Applied Ecology 1988;25:389–95. Mappes T, Grapputo A, Hakkarainen H et al. Island selection on mam- Jones KE, Bielby J, Cardillo M et al. PanTHERIA: a species-level database malian life-histories: genetic differentiation in offspring size. BMC of life history, ecology, and geography of extant and recently extinct Ecology and Evolution 2008;8:296. mammals. Ecology 2009;90:2648–2648. de Margerie E, Cubo J, Castanet J. Bone typology and growth rate: testing Jordana X, Marín-Moratalla N, Moncunill-Solé B et al. Ontogenetic and quantifying ‘Amprino’s rule’ in the mallard (Anas platyrhynchos). changes in the histological features of zonal bone tissue of ruminants: Comptes Rendus Biologies 2002;325:221–30. a quantitative approach. Comptes Rendus Palevol 2016;15:255–66. de Margerie E, Robin JP, Verrier D et al. Assessing a relationship be- Kauhala K, Soveri T. An evaluation of methods for distinguishing be- tween bone microstructure and growth rate: a fluorescent labelling tween juvenile and adult mountain hares Lepus timidus. Wildlife study in the king penguin chick (Aptenodytes patagonicus). Journal of Biology 2001;7:295–300. Experimental Biology 2004;207:869–79. Kaweblum M, Aguilar M del C, Blancas E et al. Histological and radio- Marín-Moratalla N, Jordana X, Köhler M. Bone histology as an approach graphic determination of the age of physeal closure of the distal femur, to providing data on certain key life history traits in mammals: impli- proximal tibia and proximal fibula of the New Zealand white rabbit. cations for conservation biology. Mammalian Biology 2013;78:422–9. Journal of Orthopaedic Research 1994;12:747–9. Martínez-Maza C, Alberdi MT, Nieto-Díaz M et al. Life-history traits of Klein N, Sander PM, Stein K et al. Modified laminar bone in Ampelosaurus the Miocene Hipparion concudense (Spain) inferred from bone histo- atacis and other titanosaurs (Sauropoda): Implications for life history logical structure. PLoS One 2014;9:e103708. and physiology. PLoS One 2012;7:e36907. Matsuzaki T, Saito M, Sakai A et al. Rearing of Plateau pika Köhler M, Moyà-Solà S. Physiological and life history strategies of a fossil (Ochotona dauurica) captured in Mongolia. Experimental Animals large mammal in a resource-limited environment. Proceedings of the 1998;47:203–6. National Academy of Sciences 2009;106:20354–8. Mearns EA. Descriptions of supposed new species and sub-species of Köhler M, Marín-Moratalla N, Jordana X et al. Seasonal bone growth and mammals from Arizona. Bulletin of the American Museum of Natural physiology in endotherms shed light on dinosaur physiology. Nature History 1890;2:277–307. 2012;487:358–61. Meier PS, Bickelmann C, Scheyer TM et al. Evolution of bone com- Köhler M, Herridge V, Nacarino-Meneses C et al. Palaeohistology reveals pactness in extant and extinct moles (Talpidae): exploring humeral a slow pace of life for the dwarfed Sicilian elephant. Scientific Reports microstructure in small fossorial mammals. BMC Evolutionary Biology 2021;11:22862. 2013;13:55. Köhler M, Nacarino-Meneses C, Arnold W, Stalder G, Moyà-Solà S. Millar JS, Zwickel FD. Determination of age, age structure, and mor- Defining age at maturity in the insular giant leporid Nuralagus rex tality of the pika, Ochotona princeps (Richardson). Canadian Journal from bone histology. In: Belvedere M, Mecozzi B, Amore O, Sardella of Zoology 1972;50:229–32. R (eds), Abstract Book of the XIX Annual Conference of the European Miszkiewicz JJ, van der Geer AA. Inferring longevity from advanced rib Association of Vertebrate Palaeontologists, Special Vol. 1. Benevento: remodelling in insular dwarf deer. Biological Journal of the Linnean PalaeoVertebrata, 2022, 100. Society 2022;136:41–58. 14 • Fernández-Bejarano et al. Miszkiewicz JJ, Louys J, Beck R et al. Island rule and bone metabolism Quintana J, Köhler M, Moyà-Solà S. Nuralagus rex, gen. et sp. nov., and in fossil murines from Timor. Biological Journal of the Linnean Society endemic insular giant rabbit from the Neogene of Minorca (Balearic 2020;129:570–86. Islands, Spain). Journal of Vertebrate Paleontology 2011;31:231–40. Moncunill-Solé B. Eco-evolutionary adaptations of ochotonids Reese AT, Lanier HC, Sargis EJ. Skeletal indicators of ecological spe- (Mammalia: Lagomorpha) to islands: new insights into Late Miocene cialization in pika (Mammalia, Ochotonidae). Journal of Morphology pikas from the Gargano palaeo-archipelago (Italy). Biological Journal 2013;274:585–602. of the Linnean Society 2021;132:400–13. Richardson J. Short characters of a few quadrupeds procured on Capt. Moncunill‐Solé B, Quintana J, Jordana X et al. The weight of fossil Franklin’s lake expedition. Zoological Journal 1828;3:516–20. leporids and ochotonids: body mass estimation models for the order de Ricqlès A, de Meunier FJ, Castanet J, Francillon-Vieillot H. Downloaded from https://academic.oup.com/zoolinnean/advance-article/doi/10.1093/zoolinnean/zlad112/7281581 by guest on 27 September 2023 Lagomorpha. Journal of Zoology 2015;295:269–78. Comparative microstructure of bone. In: Hall BK (ed.), Bone Matrix Moncunill-Solé B, Orlandi-Oliveras G, Jordana X et al. First approach of and Bone Specific Products, Vol. 3. Boca Raton: CRC Press, 1991, the life history of Prolagus apricenicus (Ochotonidae, Lagomorpha) 1–78. from Terre Rosse sites (Gargano, Italy) using body mass estimation and Riyahi S, Köhler M, Marín-Moratalla N, Jordana X, Quintana J. Bone paleohistological analysis. Comptes Rendus Palevol 2016;15:227–37. histology of the giant fossil rabbit Nuralagus: the trade-off between Moncunill-Solé B, Jordana X, Köhler M. Where did Mikrotia magna ori- growth and reproduction on islands. In: Köhler M, Jordana X, Marín- ginate? Drawing ecogeographical inferences from body mass recon- Moratalla N (eds), Paleontologia i evolució. I International Symposium structions. Geobios 2018;51:359–66. on Paleohistology. Sabadell: Insitut Català de Paleontologia Miquel Moncunill-Solé B, Tuveri C, Arca M et al. Tooth and long bone scaling Crusafont, 2011, 77. in Sardinian ochotonids (Early Pleistocene-Holocene): Evidence for Ruff CB. Body size, body shape, and long bone strength in modern hu- megalodontia and its palaeocological implications. Palaeogeography, mans. Journal of Human Evolution 2000;38:269–90. Palaeoclimatology, Palaeoecology 2021;582:110645. Salvador CH, Fernandez FAS. Reproduction and growth of a rare, island- Moncunill-Solé B, Arzi B, Filliquist B et al. The extinct osteoarthritic lago- endemic cavy (Cavia intermedia) from Southern Brazil. Journal of morphs from Sardinia (Prolagus sardus, Lagomorpha) reveals further Mammalogy 2008;89:909–15. evidence of life history evolution in insular domains. Palaeoworld. Schmidt-Nielsen K. Scaling: Why Is Animal Size so Important? Cambridge: https://doi.org/10.1016/j.palwor.2023.03.004. In press. Cambridge University Press, 1984. Montoya-Sanhueza G, Chinsamy A. Long bone histology of the subterra- Schoenherr AA. A Natural History of California. Berkley: University of nean rodent Bathyergus suillus (Bathyergidae): ontogenetic pattern of California Press, 1992. cortical bone thickening. Journal of Anatomy 2016;230:203–33. Skedros JG, Sybrowsky CL, Parry TR et al. Regional differences in cortical Morris PA. A method for determining absolute age in the hedgehog. bone organization and microdamage prevalence in Rocky Mountain Journal of Zoology 1970;161:277–81. mule deer. Anatomical Record 2003;274:837–50. Nacarino-Meneses C, Köhler M. Limb bone histology records birth in Smith AT. Patterns of pika (genus Ochotona) life history variation. In: mammals. PLoS One 2018;13:e0198511. Boyce MS (ed.), Evolution of Life Histories: Theory and Patterns from Nacarino-Meneses C, Jordana X, Köhler M. First approach to bone Mammals. Yale: Yale University Press, 1988, 233–56. histology and sketochronology of Equus hemionus. Comptes Rendus Smith AT, Weston ML. Ochotona princeps. Mammalian Species Palevol 2016;15:267–77. 1990;352:1–8. Nelson EW. Description of a new species of Lagomys from Alaska. Smith AT, Formozov NA, Hoffmann RS, Changlin Z, Erbajeva MA. The Proceedings of the Biological Society of Washington 1893;8:117–20. Pikas. In: Chapman JA, Flux JEC (eds), Rabbits, Hares and Pikas. Ohtaishi N, Hachiya N, Shibata Y. Age determination of the hare from annual Status Survey and Conservation Action Plan. Switzerland: International layers in the mandibular bone. Acta Theriologica 1976;21:168–71, plate II. Union for Conservation of Nature and Natural Resources, 1990, Orlandi-Oliveras G, Jordana X, Moncunill-Solé B et al. Bone histology 14–60. of the giant fossil dormouse Hypnomys onicensis (Gliridae, Rodentia) Smith AT, Johnston CH, Alves PC, Hackländer K. Lagomorphs. Pikas, from Balearic Islands. Comptes Rendus Palevol 2016;15:238–44. Rabbits and Hares of the World. Baltimore: Johns Hopkins University Padian K, Lamm ET. Bone Histology of Fossil Tetrapods: Advancing Press, 2018. Methods, Analysis, and Interpretation. Berkley: University of California Sokolov V, Yu Ivanitskaya E, Gruzdev V, Heptner V. Mammals of Russia Press, 2013. and Adjacent Regions. Lagomorphs. Moscow: Nauka, 1994. Palkovacs EP. Explaining adaptive shifts in body size on islands: a life his- Sondaar PY. Insularity and its effect on mammalian evolution. In: tory approach. Oikos 2003;103:37–44. Hecht MK, Goody PC, Hecht BM (eds), Major Patterns in Vertebrate Pallas PS. Reise durch verschiedene Provinzen des Russischen Reichs. Theil Evolution. New York: Plenum Publishing Corportation, 1977, 3. St. Petersburg: Kaiserliche Academie der Wissenschaften, 1776 [in 233–56. Russian]. Stearns SC. The Evolution of Life Histories. Oxford: Oxford University Palombo MR, Rozzi R. How correct is any chronological ordering of Press, 1992. the Quaternary Sardinian mammalian assemblages? Quaternary Studiati C. Description des fossils de la brèche osseuse de Monreale de International 2014;328-329:136–55. Bonaria près de Cagliari. In: Lamarmora AF (ed.), Voyage en Sardeigne Palombo MR, Antonioli F, LoPresti V et al. The late Pleistocene to 3. Turin: Saint Thomas, 1857, 651–704. Holocene palaeogeographic evolution of the Porto Conte area: Tacutu R, Thronton D, Johnson E et al. Human ageing genomic re- clues for a better understanding of human colonization of Sardinia sources: new and updated databases. Nucleic Acids Research and faunal dynamics during the last 30 ka. Quaternary International 2018;46:D1083–90. 2017;439:117–40. The jamovi Project. jamovi (v.2.3) [computer software]. 2022. Retrieved Peri A. Ochotona princeps. Animal Diveristy Web. 2012. https:// from: https://jamovi.org (21 March 2023, date last accessed). animaldiversity.org/accounts/Ochotona_princeps/ (21 March Trillmich F, Geißler E, Guenther A. Senescence and costs of reproduction 2023, date last accessed). in the life history of a small precocial species. Ecology and Evolution Peters RH. The Ecological Implications of Body Size. Cambridge: 2019;9:7069–79. Cambridge University Press, 1986. Wagner R. Über die Knochenbrekzie in Sardinien und die darin Pianka ER. Evolutionary Ecology, 7th edn. Austin: Publisher E.R. Pianka, gefundenen Thiere, so wie über einige andere hieher gehörige 2011. https://books.google.es/books/about/Evolutionary_ecology. Erscheinungen. Archiv für die Gesammte Naturlehre 1829;15:10–31 html?id=giFL5bonGhQC&redir_esc=y (EBook). [in German]. Quemeneur S, de Buffrénil V, Laurin M. Microanatomy of the amniote Wang Y, Li Y, Wu Z et al. Insular shifts and trade-offs in life-history traits femur and inference of lifestyle in limbed vertebrates. Biological in pond frogs in the Zhoushan Archipelago, China. Journal of Zoology Journal of the Linnean Society 2013;109:644–55. 2009;278:65–73. Life history of Prolagus sardus • 15 Waterhouse GR. Observations on the Rodentia, with a view to Xu F, Yang W, Li Y. Enlarged egg size increases offspring fitness of a frog species point out the groups, as indicated by the structure of the Crania, on the Zhoushan Archipelago of China. Scientific Reports 2019;9:11653. in this order of mammals. Magazine of Natural History, new series Young JW, Danczak R, Russo GA et al. Limb bone morphology, 1839;3:90–600. bone strength, and cursoriality in lagomorphs. Journal of Anatomy Woodward HN, Tremaine K, Williams SA et al. Growing up Tyrannosaurus 2014;225:403–18. rex: Osteohistology refutes the pygmy ‘Nanotyrannus’ and supports Young JW, Foster AD, Russo GA et al. Only the good die old? Ontogenetic ontogenetic niche partitioning in juvenile Tyrannosaurus. Science determinants of locomotor performance in Eastern cottontail rabbits Advances 2020;6:eaax6250. (Sylvilagus floridanus). Integrative Organismal Biology 2022;4:obab037. Downloaded from https://academic.oup.com/zoolinnean/advance-article/doi/10.1093/zoolinnean/zlad112/7281581 by guest on 27 September 2023