Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2025 Sep-Oct;247(3-4):765-789.
doi: 10.1111/joa.14186. Epub 2025 Feb 5.

Osteohistology of the unusually fast-growing theropod dinosaur Ceratosaurus

Riley Sombathy  1   2 , Patrick M O'Connor  2   3   4   5 , Michael D D'Emic  6   7
Affiliations

Osteohistology of the unusually fast-growing theropod dinosaur Ceratosaurus

Riley Sombathy et al. J Anat. 2025 Sep-Oct.

Abstract

Ceratosaurus is a large-bodied non-avian theropod dinosaur known from the Upper Jurassic Morrison Formation of North America and is remarkable both for its exceptionally fast annual growth rate and its status as the only theropod currently known with postcranial osteoderms. We describe the osteohistology of three hind limb bones, two dorsal ribs, and one osteoderm representing four individuals of Ceratosaurus. In addition to describing the tissues of these bones, we compared the annual growth rates from three individuals in our sample to those of five other ceratosaurians. We fit seven growth models to two of the specimens in our sample and compared the results of the best-fit model(s) to those of two other ceratosaurians (Masiakasaurus knopfleri and Majungasaurus crenatissimus) for which sufficient growth data were available. The bone tissue of hind limbs in Ceratosaurus is highly vascularized, with dense plexiform or reticular vascular complexes and alternating strips of parallel or woven-fibered matrix. Few lines of arrested growth were recorded in hind limbs prior to specimens achieving asymptotic body size. Both sampled dorsal ribs are highly remodeled, with only small portions of primary bone visible in each section, revealing parallel-fibered bone with sparse primary osteons. Both dorsal ribs contain numerous lines of arrested growth throughout the cortex that allowed for more accurate estimates of individual age when paired with the data from hind limbs. The osteoderm is composed of a core of large Haversian canals and a perimeter of lamellar bone with dense Sharpey's fibers along the internal surface of the bone. Multiple LAGs are also present within the lamellar bone along the exterior margins. Maximum annual growth rates in Ceratosaurus were on average nine-fold faster than those of other ceratosaurians. Our sample lacks data from juveniles so confidence in inferred growth models is limited. Thus, to begin to constrain Ceratosaurus growth patterns, we averaged the results of all models that possessed an Akaike Information Criterion score corrected for small sample size (AICc) within 10 of the lowest scoring model. We found that the monomolecular model exhibited the lowest AICc value, with the von Bertalanffy and Gompertz models possessing AICc values within 10 units of it. In contrast, the logistic and Gompertz models were confidently selected for Masiakasaurus and Majungasaurus, respectively. Irrespective of growth model, maximum relative annual growth rates for Ceratosaurus were several-fold greater than those of Masiakasaurus and Majungasaurus. Both histological and growth model estimates of life history support an evolutionary trend towards more prolonged development in Ceratosauria through evolutionary time.

Keywords: biology; growth; life history; paleohistology; paleontology.

PubMed Disclaimer

Figures

FIGURE 1
FIGURE 1
Geographic map of the United States of America and state maps of Utah and Colorado, with black stars and labels indicating localities from which Ceratosaurus specimens included in this work were collected (a). Skeletal reconstruction of Ceratosaurus specimens next to relevant localities with those bones used for histological sampling highlighted in red. Stratigraphic context for Morrison Formation, adapted from Maidment & Muxworthy (2019), with the approximate position of localities marked with black stars (b). ABQ, Agate Basin Quarry; CLDQ, Cleveland‐Lloyd Dinosaur Quarry; DMQ, Dry Mesa Quarry; FPA, Fruita Paleontological Area; Ma, Millions of years ago; Tid, Tidwell Member. For A1‐C6 abbreviations, see Maidment and Muxworthy  (2019). Skeletal reconstruction from Scott Hartman, see Supplemental File—Data S1 for Phylopic license.
FIGURE 2
FIGURE 2
Osteohistology of the femur (BYU 725–5133) of Ceratosaurus from the Dry Mesa Quarry, Colorado. Dashed white lines (d and e) represent approximate transition zones between different bone matrix classes. Cross section using brightfield (BF) light microscopy with schematic in the lower left corner of figure section shows a hypothetical full cross section of a Ceratosaurus femur based on the complete cross section of an Allosaurus femur (TATE v3587); diagonal lines represent estimated position of sampled bone in section (a). Longitudinal section using cross‐polarized light (XPL) microscopy (b). Magnified portion of longitudinal section in XPL (c). Remodeled bone in XPL (d). Endosteal lamellae and vascular canals in XPL (e). Woven fiber bone with plexiform vascular canals in the anterior region in XPL (f). Parallel‐fibered bone with plexiform vascular canals in XPL (g). Scale bar equals 5 mm in (a, b), 1 mm in (c–e), and 500 μm in (f–g). A‐ Anterior, CC, circumferential canals; EL, endosteal lamellae; L, lateral; LC, longitudinal canals; M, medial; MC, medullary cavity; OL, osteocyte lacuna; P, posterior; PFB, parallel‐fibered bone; PO, primary osteon; RB, remodeled bone; RC, radial canals; ST, stain; WFB, woven‐fibered bone.
FIGURE 3
FIGURE 3
Osteohistology of the femur (BYU 881–12893) of Ceratosaurus from the Agate Basin Quarry, Utah. Dashed white lines (d and e) represent approximate transition zones between different bone matrix classes. Cross section using brightfield light (BF) microscopy, schematic in the lower left corner of figure section shows a hypothetical full cross section of a Ceratosaurus femur based on the complete cross section of an Allosaurus femur (TATE v3587); diagonal lines represent estimated position of sampled bone present in section (a). Longitudinal section using cross‐polarized light (XPL) microscopy (b). Magnified portion of longitudinal section in XPL (c). Resorption room in XPL (d). Remodeled bone along the anterior periosteal border in XPL (e). Parallel‐fibered bone with plexiform vascular canals in the medial region of the cortex in XPL (f). Nine LAGs along the periosteal border (g). Scale bar equals 5 mm in a–b and 1 mm in c–g. Arrow heads and red lines denote LAGs whereas blue dashed lines indicate estimated connections between visible LAGs. A, anterior; CC, circumferential canals; EL, endosteal lamellae; HC, Haversian canal; L, lateral; LC, longitudinal canals; M, medial; MC, medullary cavity; P, posterior; PO, primary osteon; RC, radial canals; RB, remodeled bone; ST, stain.
FIGURE 4
FIGURE 4
Osteohistology of a dorsal rib (BYU 881–12893) of Ceratosaurus from the Agate Basin Quarry, Utah. Dashed white lines (b) represent approximate transition zones between different bone matrix classes. Cross section using cross‐polarized light (XPL), schematic in the lower right corner of figure section shows the interpreted anatomical borders of BYU 881–12893; diagonal lines that represent estimated position of sampled bone present in section (a). Periosteal margin of bone showing multiple LAGs and remodeled cortex in XPL (b). Closer image of EFS in XPL (c). Scale bar equals 5 mm in (a), 1 mm in (b), and 500 μm in (c). Arrow heads denote LAGs. A, anterior; HC, Haversian canal; L, lateral; M, medial; P, posterior; PO, primary osteon; RB, remodeled bone; ST, stain.
FIGURE 5
FIGURE 5
Osteohistology of the tibia (MWC 1) of Ceratosaurus from the Fruita Paleontological Area, Colorado. Dashed white lines (d–f) represent approximate transition zones between different bone matrix classes. Reconstructed cross section using brightfield light (BF) microscopy; schematic in the lower left corner of figure section shows a reconstructed cross section of MWC 1 based on the complete cross section of a Majungasaurus crenatissimus tibia (DMNH EPV.135993); diagonal lines represent estimated position of sampled bone present in section (a). Longitudinal section using cross‐polarized light (XPL) microscopy (b). Magnified portion of longitudinal section in XPL (c). Secondary remodeling along anterior periosteal border in XPL (d). Plexiform vascular canals and transition between parallel‐fibered and woven‐fibered bone in XPL (e). Strip of reticular vascularity and woven‐fibered bone along the anterior region of the cortex in XPL (f). Five inner LAGs and EFS in BF (g). Scale bar equals 10 mm in a and 1 mm in (b–g). Arrow heads and red lines denote LAGs, blue lines denote hypothesized connections between visible LAGs. A, anterior; CC, circumferential canals; HC, Haversian canal; L, lateral; LC, longitudinal canals; M, medial; P, posterior; PFB, parallel‐fibered bone; RB, remodeled bone; RC, radial canals; WFB, woven‐fibered bone.
FIGURE 6
FIGURE 6
Osteohistology of a dorsal rib (UMNH VP 5278) of Ceratosaurus from the Cleveland‐Lloyd Dinosaur Quarry, Utah. Dashed white lines (d and e) represent approximate transition zones between different bone matrix classes. Cross section using brightfield light (BF) microscopy; schematic in the lower right corner of figure section shows reconstructed cross section of UMNH VP 5278 based on the contours of the bone; diagonal lines represent estimated position of sampled bone present in section (a). Cross section using cross‐polarized light (XPL) microscopy (b). Remodeled bone with Haverisian canals transistioning into lamellar bone peripherally with interspersed LAGs along lateral periosteal border in XPL (c). Endosteal lamellae along the deep anterior border with large medullary trabeculae in XPL (d). LAGs forming an EFS along the lateral periosteal border in BF (e). Primary bone with LAGs in the deep medial cortex and the EFS along the medial periosteal border in BF (f). Scale bars are 5 mm in (a, b) and 1 mm in (c–f). Arrow heads and red lines denote LAGs whereas blue lines indicate hypothesized connections between LAGs. A, anterior; EL, endosteal lamellae; HC, Haversian canal; L, lateral; LB, lamellar bone; LM, lamellae; M, medial; MC, medullary cavity; P, posterior; PO, primary osteon; RB, remodeled bone.
FIGURE 7
FIGURE 7
Pictures of Ceratosaurus osteoderm dermis patterns. Surface patterns on osteoderms of Ceratosaurus to indicate the internal (i.e., body facing) surface of the osteoderm. UMNH VP 5278 (UUVP 80) (a), UMNH VP 5278 (UUVP 67826) (b), UMNH VP 5278 (UUVP 677) (c). Scale bars equal 1 mm. CHT, Cross‐hatched Texture; NF, Nutrient Foramina.
FIGURE 8
FIGURE 8
Osteohistology of osteoderm (UMNH VP 5278) of Ceratosaurus from the Cleveland‐Lloyd Dinosaur Quarry, Utah. Cross section using cross‐polarized light (XPL) microscopy; schematic on the upper left of figure section shows cross section of UMNH VP 5278; diagonal lines represent estimated position of sampled bone present in section (a). A large canal running through the cortex of the bone in XPL (b). Greyscale image of the bone cortex in XPL showing structural fibers running at orthogonal angles (c). Haversian canals along periosteal border in XPL (d). Haversean canals in the cortex transistioning to lamellar bone seperated by eight LAGs forming an EFS in brightfield light (BF) (e). Sharpey's fibers running through the cortext along the internal (i.e., body facing) periosteal border in XPL (f). Same image as (f) converted to gray scale to visulaize the Sharpey's fibers (g). Scale bar equals 5 mm in (a), 1 mm in (b–e), and 500 μm in (f, g). Arrow heads denote LAGs. DEEP, deep surface; HC, Haversian canal; RB, remodeled bone; SF, structural fibers; ShF, Sharpey's fibers; SUP, superficial.
FIGURE 9
FIGURE 9
Individual and mean growth curves generated from average bootstrapped values of the Average (a), Monomolecular (b), von Bertalanffy (c), and Gompertz (d) models for Ceratosaurus, Gompertz model Masiakasaurus knopfleri (e), and logistic model for Majungasaurus crenatissimus (f). DME, Developmental Mass Extrapolation. Silouettes from Scott Hartman, see Supplemental File—Data S1 for Phylopic liscene.
FIGURE 10
FIGURE 10
Bar graphs of body mass for BYU 725–5133 (femur), BYU 881–12893 (femur), and MWC 1 (tibia). Colors and patterns of bars correspond with bone matrix and canal types respectively. LAGs represented by horizontal lines and arrow heads. Missing data represented by diagonal lines. Endosteal border designated by arrows and text in figure. For further explanation see key in figure. DME, Developmental Mass Extrapolation; EFS, External Fundamental System; LAG, Line of Arrested Growth.
FIGURE 11
FIGURE 11
Bar graphs of the maximum percentage of asymptotic body mass gained in a year (%) for specimens of Ceratosaurus, Eoabelisaurus, Majungasaurus, Masiakasaurus, Quilmesaurus, and Vespersaurus (a). Growth data for Eoabelisaurus, Majungasaurus, Masiakasaurus, Quilmesaurus, and Vespersaurus sourced from supplemental files of D'Emic et al., , and are available in (Supplemental File—Data S1). Red lines represent individual annual growth rates, black lines represent the average annual growth rate for taxa from which multiple specimens were available. Numbers refer to specimen numbers found in Supplemental File—Data S1. Cross section of the femur of BYU 881–12893 showing the large growth zone in pink, missing data represented by diagonal lines, LAGs as solid red lines and hypothesized connections between LAGs as dashed blue lines (b). Silouette from Scott Hartman, see Supplemental S1 for phylopic liscene. LAG, Line of Arrested Growth.
See this image and copyright information in PMC

References

    1. Amprino, R. (1947) La structure du tissu osseux envisagée comme expression de différences dans la vitesse de l'accroissement. Archives de Biologie, 58, 315–330.
    1. Aureliano, T. , Ghilardi, A.M. , Fonseca, P.H.M. , Martinelli, A.G. & Marinho, T.S. (2024) The evolution and diversification of growth strategies in abelisauroid theropods. Journal of Vertebrate Paleontology, 43, e2298395. Available from: 10.1080/02724634.2023.2298395 - DOI
    1. Barta, D.E. , Griffin, C.T. & Norell, M.A. (2022) Osteohistology of a Triassic dinosaur population reveals highly variable growth trajectories typified early dinosaur ontogeny. Scientific Reports, 12, 17321. Available from: 10.1038/s41598-022-22216-x - DOI - PMC - PubMed
    1. Brassey, C.A. , O'Mahoney, T.G. , Chamberlain, A.T. & Sellers, W.I. (2018) A volumetric technique for fossil body mass estimation applied to Australopithecus afarensis . Journal of Human Evolution, 115, 47–64. Available from: 10.1016/j.jhevol.2017.07.014 - DOI - PubMed
    1. Broeckhoven, C. & du Plessis, A. (2022) Osteoderms as calcium reservoirs: insights from the lizard Ouroborus cataphractus . Journal of Anatomy, 241, 635–640. Available from: 10.1111/joa.13683 - DOI - PMC - PubMed

Publication types

MeSH terms

LinkOut - more resources