Azahara Salazar-Fernández ^1*, José Miguel Carretero ^1,2,3, Laura Rodríguez^1,4 & Rebeca García-González¹

¹ Laboratorio de Evolución Humana. Universidad de Burgos. Edificio I+D+i/CIBA, Plaza Misael Bañuelos s/n, 09001 Burgos, Spain

²Centro UCM-ISCIII de Investigación sobre Evolución y Comportamiento Humanos, Avda. Monforte de Lemos 5 (Pabellón 14), 28029 Madrid, Spain

³Unidad Asociada de I+D+i al CSIC Vidrio y Materiales del Patrimonio Cultural (VIMPAC).

⁴Departamento de Biodiversidad y Gestión Ambiental. Universidad de León. Facultad de Ciencias Biológicas y Ambientales. Campus de Vegazana. Avda. Emilio Hurtado s/n 24071 León, Spain

* Corresponding author: azahara_sf@hotmail.com

https://doi.org/ 10.54062/jb.4.3.3

Abstract

This study investigates the morphological changes of the coracoid process and its metaphyseal surface at the glenoid-coracoid interface, aiming to characterize these transformations across different maturity stages. A total of 26 coracoid epiphyses and 48 coracoid metaphyses from a skeletal sample excavated at the medieval Dominican Convent of San Pablo in Burgos, Spain, were analysed. The sample was 3D scanned and divided into four distinct maturity groups based on developmental criteria. Utilizing three-dimensional geometric morphometric analysis, two different 3D template configurations were created to capture the shape changes during the development of the coracoid process and its metaphysis, respectively. The analysis included Shape and Form space analysis, allowing for an evaluation of maturity transitions and their interrelationship with size. The research identified the first principal component, both in Shape and Form Spaces Analysis, as the most suitable for elucidating ontogenetic changes in both shape and form. The findings demonstrate a clear, coordinated progression of the coracoid process and coracoid metaphysis morphology from childhood to adolescence, indicating that the morphological changes of the articular surfaces of the coracoid epiphysis influence the changes in the metaphyseal surfaces. By correlating these changes, the study infers potential new methods for estimating skeletal maturity in dry bone contexts. Moreover, this research emphasizes the significance of both size and shape variations in assessing skeletal maturity, with shape differentiation playing a more prominent role in later developmental stages than size. Ultimately, this study contributes to a deeper understanding of skeletal growth patterns and offers valuable insights for refining age estimation methodologies, particularly in anthropological and forensic contexts.

Keywords: Coracoid process, metaphysis, geometric morphometrics skeletal maturity.

Introduction

Constructing age distributions from skeletal remains is a critical aspect of anthropological research. The most reliable methods for estimating the age of non-adult individuals are based on dental development (Liversidge, 2008). However, suitable dental material is not always available, making it necessary to explore alternative methods. In such cases, the diaphyseal growth in the length of long bones is considered a valuable approach for age estimation, particularly in preadolescent individuals (Cardoso et al., 2014).

Several predictive equations have been developed based on different modern samples, each with varying growth patterns, body sizes, and proportions, to estimate age from diaphyseal lengths (Cowgill et al., 2010; Rissech et al., 2008, 2013; Lopez-Costas et al., 2012; Cardoso et al., 2014; Stull et al., 2014). Additionally, though less frequently, some studies have also incorporated measurements of the clavicle and scapula, focusing on increases in their length and width. (Rissech and Black, 2007; Cardoso et al., 2017). In all cases, choosing the correct predictive equation for age estimation is crucial to obtain an accurate estimation. An effective way to choose the appropriate equation is by assessing bone size in relation to its maturity or skeletal development, as is commonly done in clinical settings. Although, skeletal age can be influenced by factors such as sex, nutrition, metabolic and genetic conditions, and social circumstances (Cavallo et al., 2021), generally, while individuals showing an advanced skeletal development, while tall for their chronological age, tend to be short for their skeletal ages, individuals showing a delayed skeletal development, are short for their chronological age but tall for their skeletal age (Martin et al., 2011; Bayley 1946). Therefore, the combined assessment of ages obtained through different prediction equations based on long bones and those derived from skeletal development is essential for achieving an accurate estimation.

The process of maturity involves both quantitative and qualitative changes that transform an undifferentiated, immature state into a highly organized, specialized, and mature state (Roche, 1992). The assessment of skeletal maturity is based on a series of indicators that are discrete events or recognizable states within the continuous process of maturation (Cameron, 2004). Several methods have been developed based on maturity indicators that describe the sequence of the onset of ossification in the epiphyses of the long bones, the changes in shape and size of the epiphyses, the chronology of the epiphyseal union, and the percentage of adult size achieved (Eveleth and Tanner 1990; Humphrey 2003).

In bioarcheological contexts is very common the use of standards charting the chronology of unions of epiphyses to diaphyses for age estimation (Stevenson 1924; Todd1930; Stewart 1934; Coqueugniot and Weaver 2007; Schaefer and Black 2007; Cardoso2008, b; Cardoso and Ríos 2011; Cardoso et al. 2014). However, these standards are restricted to a concrete period of life, as their utility is limited to pre-adolescents and adolescents.

This has led several authors to explore the application of indicators based on changes in the shape and size of the epiphyses, which can be tracked throughout the entire development period (Conceição and Cardoso, 2011; García-González et al., 2019). Additionally, there are studies focused on defining maturity states based on metaphyseal changes (García-González et al., 2024a; Salazar-Fernández et al., under review). All of these have proven useful for assessing the maturity status of archaeological individuals, as well as for age estimation. Thus, it is reasonable to continue adapting or developing methods based on maturity indicators that can be applied through direct observations of dry bones. In this context, the scapula is a strong candidate. Unlike long bones, which grow primarily in length through two metaphyses (proximal and distal), the scapula develops through multiple metaphyseal areas. It has at least seven secondary ossification centres and one primary ossification centre (Cardoso, 2008; Ogden & Phillips, 1983). These centres include the articular base of the glenoid-coracoid interface, the inferior angle, the vertebral border, the acromial process, the glenoid rim, and the apex and angle associated with the coracoid process (Kothary et al., 2014; Scheuer & Black, 2004). Most of these ossification centres appear, form, and fuse during adolescence.  The scapula, like other flat bones, is highly fragile. Consequently, in archaeological contexts, only its thicker parts, such as the glenoid cavity, coracoid process, or spine, are often recovered. This limits the ability to estimate age at death based on the main dimensions of the scapula. Existing studies on non-adults, such as those by Ogden & Phillips (1983) have concentrated on the global ossification of the scapula and how various pathologies and traumas impact its development, as observed through radiographic data. Similarly, Kothary et al. (2014), examined the presence or absence of the ossification centres that compose only the glenoid and the glenoid-coracoid interface, using a sample analysed through magnetic resonance imaging (MRI). Unfortunately, radiographic studies on the scapula are limited due to visualization challenges within the appendicular skeleton (Ogden & Phillips, 1983). Furthermore, such studies are difficult to extrapolate to dry bone samples. Yet, García-González et al. (2024a), have shown that developmental changes in the scapular glenoid cavity can be a valuable tool for estimating age at death. Although this study focused on the maturity changes at the fusion site of the coracoid process, there has not been an extensive assessment of its overall developmental process. Therefore, more exhaustive research on the ontogenetic changes associated with scapular ossification is necessary to better understand its developmental complexity in dry bone samples. Most of the secondary ossification centres of the scapula appear, form, and fuse during adolescence (Scheuer et al., 2004). However, the coracoid process undergoes a prolonged process of skeletal maturation, from its initial appearance to its fusion in adolescence, making it an epiphysis with a longer developmental timeline compared to other scapular epiphyses. While the coracoid process is often considered a primary ossification centre due to its potential formation during prenatal development, unlike secondary centres that always appear postnatally, it behaves more like a secondary centre as it fuses postnatally (Scheuer & Black, 2000). Therefore, given its prolonged developmental period and the fact that it is recognizable early in this process, the coracoid process can be useful for skeletal age estimation.

In this study, the morphological changes of both the coracoid epiphysis and its metaphyseal interface on the glenoid-coracoid interface are analyzed using geometric morphometrics, a method well-suited to analyze and quantify the shape of biological structures through spatial coordinates (Bookstein, 2018; Mitteroecker et al., 2013, 2022; Mitteroecker & Gunz, 2009). Unlike traditional techniques that measure lengths, areas, or volumes, geometric morphometrics allows for the consideration of the geometry and spatial relationship between landmarks in a structure (Bastir et al., 2013; Gunz & Mitteroecker, 2013; Zelditch, 2004).

Materials and methods

Sample Selection and Group Classification

This study investigated 26 coracoid epiphyses, and 48 metaphyseal surfaces for the fusion of the coracoid on the glenoid-coracoid interface from non-adult skeletal remains excavated at the Dominican Convent of San Pablo in Burgos, Spain. The burials at the Convent cover a wide chronological range from the 13th to the late 19th century, but this research specifically focuses on remains dated from the 14th to the 18th centuries (García-González et al., 2024b). All individuals were carefully selected to exclude any exhibiting deformation or visible pathology.

Ages of the non-adult individuals from this collection were estimated based on the calcification and formation of their dental crowns and roots (García-González et al., 2024). Mineralization stages for each tooth type were analyzed using 3D volume renderings obtained from CT scans. The developmental stages of the permanent dentition were scored according to the method established by Demirjian et al. (1973) and subsequently converted into age estimates using the adjusted prediction data proposed by Liversidge et al. (2006). For deciduous teeth, the method developed by Liversidge and Molleson was applied (2004).

The total sample was divided into four maturity groups based on both the pattern of human growth and development by Bogin, (2021), and the specific pattern of scapular development as described by Kothary et al. (2014), Ogden and Phillips, (1983), and Scheuer et al. (2004). Radiographic evidence indicates that while the coracoid process may emerge prenatally, it is typically first observed at around 3 months of age postnatally, and is consistently present by the end of the first year of life. The coracoid process begins a marked expansion around 2 years of age, establishing a true bipolar growth region between the glenoid and the glenoid-coracoid interface, which reflects the developmental independence of the main scapular body and the coracoid (Figure 1). Consequently, our initial maturity group is set at 3 years of age, a stage at which the coracoid process has attained clear and easily recognizable morphological features.

Figure 1: Developmental sequence of coracoid and glenoid-coracoid interface in antero-lateral view (upper) and lateral view (bottom)

a: Coracoid and glenoid-coracoid between birth and 3 years old, b: appearance and initiation of the coracoid ossification centre, c: the coracoid and its metaphysis in the glenoid-coracoid interface are morphologically recognizable, d: extension of the subcoracoid and rotation of the coracoid metaphysis in the glenoid-coracoid interface towards anterior, e: the coracoid process completes formation before its fusion with the subcoracoid.

This group encompasses the entirety of childhood, concluding at 7 years of age. During this developmental stage, the sole ossification event observed in the scapula pertains to the continued formation and maturation of the coracoid process. The second maturity group comprises individuals aged 7 to 10 years, representing the prepubertal and pubertal stages. Specifically, between 8 and 10 years, the first secondary ossification centre of the scapula, known as the subcoracoid, emerges. This centre is located dorsally to the base of the coracoid process and is responsible for the formation of the upper third of the glenoid cavity. The third group, consisting of individuals aged 10 to 14 years, represents early adolescence. The fourth group, including those from 14 to 16 years old, represents intermediate adolescence. This period concludes with the final fusion of the subcoracoid and coracoid processes to the scapular body. Initially, fusion occurs between the subcoracoid and coracoid processes, and once united, they briefly fuse with the body of the scapula. These maturity stages, along with their most probable age ranges of occurrence and the number of individuals used for each are depicted in Table 1.

Table 1. Maturity groups for the development of coracoid and glenoid-coracoid interface.

Data processing and 3D geometric morphometric Analysis.

All individuals were scanned with a 3D structured light white surface scanner, which employs trigonometric triangulation to capture surface light patterns, creating a precise 3D representation of each subject’s surface. The device utilized was the EinScan Pro (Shining 3D Tech. Co., Ltd., Hangzhou, China), operated in fixed mode with an automated turntable, and controlled via EinScan Pro software (Solid Edge SHINING 3D Edition). Each scan achieved an accuracy of 0.04 mm and was obtained from 30 different angles, ensuring comprehensive 360-degree coverage. Following acquisition, all scans were aligned and merged to produce a cohesive 3D model. These models were then exported as. ply files to Meshmixer software (Co. Autodesk, Meshmixer, USA) for post-processing, addressing minor defects such as closing surface holes.

Two 3D template configurations were developed to characterize the morphology of the coracoid process and the metaphyseal coracoid surface of the glenoid-coracoid interface. These templates were applied across all the corresponding three-dimensional samples. The coracoid process template consists of 10 fixed landmarks and 77 curved semilandmarks, while the metaphyseal coracoid surface template comprises 4 fixed landmarks and 20 curved semilandmarks (Figure 2) (Table 2 and 3).

Additionally, to address the uncertainty regarding landmark placement along both surfaces, semilandmarks were adjusted along their respective curves in relation to the fixed landmarks to reduce bending energy. This adjustment process was conducted first for each specimen in relation to a randomly selected template and then again against the average configuration of the sample (Bastir et al., 2013; García-Martínez et al., 2020; Mitteroecker & Gunz, 2009). Both templates were developed using Stratovan Checkpoint (Stratovan Corporation, 2020) by the same researcher (AS-F) to minimize interobserver measurement errors. Subsequently, this software was employed to apply each template configuration to its corresponding sample, further optimizing the alignment between the sample and the template and ensuring reduced bending energy (Gunz and Mitteroecker, 2013; Slice, 2006). To proceed with the analysis, the raw coordinates of landmarks and semilandmarks were imported into MorphoJ (Klingenberg, 2011) to conduct a General Procrustes Analysis (Rohlf & Slice, 1990; Slice, 2006). This analysis employed a Procrustes superimposition technique to minimize discrepancies between homologous landmarks by adjusting their location, rotation, and scale. Therefore, size was quantified using a scaling factor known as centroid size, which is defined as the square root of the sum of the squared distances from all landmarks to their centroid—the mean of the x, y, and z coordinates for all landmarks (Klingenberg, 2016; Mitteroecker et al., 2013). This procedure transformed the raw coordinates into Procrustes shape coordinates, capturing shape information that reflects size-related differences, while excluding overall size (Bookstein, 1991; Rohlf and Slice, 1990). The Procrustes coordinates were subsequently exported to PAST 4.13 software, where a principal components analysis was performed to examine shape variations among the different maturity groups (García‐Martínez et al., 2020; Gómez‐Olivencia et al., 2018; Harvati et al., 2019; Harvati et al., 2024; Karakostis & Harvati, 2021; Mitterœcker & Schæfer, 2022; Morley et al., 2022).

The broken stick method was utilized to determine which principal components were statistically significant (Frontier, 1976). Considering that the development process not only consists of an increase in size (growth) but also involves a maturation process (changes in shape), and that these two processes are closely related, a form analysis was also conducted by incorporating the natural logarithm of centroid size (logCS) as an additional variable alongside the Procrustes shape coordinates (Mitteroecker et al., 2004). This approach guarantees that for isotropic landmark variation, the distribution in the size-and-shape space remains isotropic. The resulting expanded dataset, which included this additional variable, underwent a principal components analysis, allowing the distances within this broader space to be interpreted as a measure of form differences (Freidline et al., 2013; Mori and Harvati, 2019). To evaluate the degree of covariation between two sets of shape variables (the metaphysis of the coracoid glenoid-coracoid interface shape variables and the coracoid process epiphysis shape variables), we performed a two-block Partial Least Squares (2B-PLS) analysis. This method characterizes the multivariate relationships between the two shape variable blocks by identifying pairs of latent variables that maximize the covariance between them (Klingenberg, 2010; Klingenberg & Marugán-Lobón, 2013; Rohlf & Corti, 2000).

Results

Shape and Form Variation in the Coracoid Process Across Maturity Stages.

Analysis of relative warps across the entire sample yielded 25 principal components. Applying the broken stick criterion revealed that only the first four PCs were statistically significant, accounting for 61.45% of the total variance (Figure 3).

Figure 3: Principal component values obtained from the morphological analysis of the coracoid process. Broken stick values are indicated by a pink dashed line.

PC1 and PC2 explained 22.28% and 17.71% of the variability, respectively, representing a combined 39.99% of the total variance.  Within the statistically significant variance, PC1 and PC2 together accounted for 65.07%. Therefore, our subsequent description of shape changes in the proximal metaphyseal surface during development focuses primarily on PC1 and PC2 (Figure 4).

Figure 4: (A) Shape space analysis of the coracoid process, where the red line represents the mean centre of each group. (B) Form space analysis of the coracoid process, similarly, illustrating the red line as the mean centre across groups.

Shape changes along PC1 primarily reflect the transformation of the coracoid process from a short, less curved form with a rounded and flat articular base at lower values, to a significantly more curved and elongated coracoid with a quadrangular-shaped articular base with a well-defined articular area for the subcoracoid at higher values (Figure 4a). In contrast, along PC2, differences are focused mainly on the articular base. While the morphology of the coracoid body is similar across PC2, the negative values describe a slightly more curved coracoid body compared to the positive values. However, the articular base is flatter and has a less defined articular area for the subcoracoid in the negative values compared to the positive values (Figure 4a).

The first axis (PC1) shows a clear progression of groups, beginning with the youngest group, which is located at the lowest values of the axis, while individuals of more advanced ages from the skeletal sample are positioned at higher values. Thus, in the Shape Space Analysis, the first maturity group is found at the most negative values of the axis, followed by the second maturity group, which has some of its individuals in the negative section, overlapping with the highest values of the first maturity group. The other half of the second maturity group is positioned in the positive values of the axis, overlapping with the lower values of the third maturity group. Both the third and fourth maturity groups are located in the positive values, with the fourth group occupying the highest values on the axis.

The first and second principal components of the Form Space analysis account for 86.6% and 3.39% of the variance, respectively (Figure 4b). The distribution of the groups along the first principal component in the Form Space Analysis is similar to that observed in the Shape Space Analysis, but the third maturity group shows a greater degree of overlap with the fourth maturity group (Figure 4b). This suggests that the size changes are more pronounced between the first and second maturity groups, as well as between the second and third, compared to the changes between the third and fourth maturity groups.

In contrast, along PC2 (in both the Shape and Form Space Analyses) individuals from all maturity groups are distributed across both axes, showing no clear separation between them.

Given the observed shape and form changes across the positive and negative values of PC1, as well as the consecutive progression of the maturity groups along this axis, this principal component can be interpreted as representing the ontogenetic sequence. Therefore, along PC1, the shape changes corresponding to the mean values of each maturity group are as follows. In the first group, the coracoid is short and slightly curved, with a rounded and flat articular base. However, as we progress from the first to the second group, the coracoid elongates and becomes wider, while its articular base begins to lose its rounded shape. The medial portion of the base expands, and the lateral border becomes more distinctly marked for the articular area of the subcoracoid (Figure 5).

Figure 5: Ontogenetic sequence illustrating shape transformations of the coracoid process across different perspectives. Views categorized under A represent morphological changes in the coracoid process derived from mean values of maturity groups along the primary axis (PC1) in the geometric morphometric analysis. Views categorized under (B) present 3D digital reconstructions from each age cohort within the sample, serving as a comparative framework for the morphometric results shown in category (A).

In the transition from the second to the third maturity group, the coracoid body not only elongates but also appears more curved, with a more pronounced hook-like shape. The articular base elongates medially to laterally, and the articular portion of the subcoracoid becomes more expanded. The anterior part of the medial edge begins to extend more pronouncedly. Finally, in the fourth maturity group, the apex of the coracoid elongates considerably and becomes more pointed. The articular base of the coracoid has angular medial borders, and the portion of the base that articulates with the subcoracoid extends, creating a small notch on its anterior aspect that reflects the morphology of the glenoid notch positioned between the glenoid cavity and the metaphyseal interfaces of the subcoracoid and coracoid (glenoid-coracoid interface) (Figure 5).

Shape and Form Variation in the Metaphyseal Coracoid Surface of the Glenoid-Coracoid Interface Across Maturity Stages.

Principal component analysis of the entire sample yielded 47 components, of which the first five were statistically significant (based on the broken stick criterion), accounting for 72.03% of the total variance (Figure 6). PC1 and PC2 explained 25.71% and 16.16% of the variability, respectively, totalling 41.87%. When considering only the variance captured by the statistically significant PCs, PC1 and PC2 together represent 58.13% of that variance. Thus, subsequent shape analysis of the proximal metaphyseal surface during development will focus on PC1 and PC2.

Figure 6: Principal component values derived from the morphological analysis of the coracoid metaphysis within the glenoid-coracoid interface. Broken stick model values are represented by a pink dashed line.

In the shape space analysis, PC1 reflects the morphological transformation of the glenoid-coracoid metaphysis, shifting from a short, rounded, and flat shape at lower values to a quadrangular surface tilted towards the antero-inferior plane at higher values. In contrast, PC2 captures changes in the proportions of the metaphyseal interface, transitioning from a narrower yet longer metaphysis at more positive values to a wider-than-long metaphysis at negative values.

PC1 in the Shape Space Analysis clearly separates individuals by maturity group, with younger individuals concentrated at lower values and older individuals at higher values. The youngest individuals (the first maturity group) are situated at the most negative values, while maturity group 2 is distributed around zero, with roughly half of its members in the negative range and half in the positive. Maturity group 4 is located in the positive values and maturity group 5 is located at the highest positive values. The clearest separation is observed between maturity groups 1 and 2, while groups 4 and 5 show some small overlap (Figure 7a). The Form Space Analysis reveals that the first principal component accounts for 80% of the variance, while the second principal component accounts for 7.38% (Figure 7b). Compared to the Shape Space Analysis, PC1 shows greater overlap among the last three maturity groups. The first group is situated at the lower values of the PC1 axis, while the second group has most of its values in the negative part of the axis.

Figure 7: (A) Shape space analysis of the coracoid metaphysis within the glenoid-coracoid interface. The red line across groups indicates the mean centre of each group. (B) Form space analysis of the coracoid metaphysis in the glenoid-coracoid interface, with the red line representing the mean centre across groups.

The positive values of the second group overlap with the lower values of the third group, and the third and fourth groups demonstrate the greatest degree of overlap.

As with the analysis of the coracoid process, the distribution of individuals along PC2 shows no clear separation between maturity groups in either the Shape or Form space analysis (Figure 7). Therefore, once again, the clear progression of maturity groups along PC1 and their associated shape changes, indicates that PC1 is the axis that explains the ontogenetic sequence of the sample. Along PC1, the changes in the shape of the coracoid methapysis at the glenoid-coracoid interface are as follows, In the first maturity group, the metaphyseal surface is characterized by being short, flat, and with a rounded medial border (Figure 8). However, in the second maturity group, the medial border becomes more expanded. Additionally, the posterior border begins to elevate and develop superiorly, beginning to twist slightly toward the infero-anterior plane. Upon reaching the third maturity group, the medial border develops a quadrangular morphology. At this stage, a full torsion of the entire metaphysis toward the infero-anterior plane becomes evident.

Transitioning to the fourth maturity group, there is a pronounced mediolateral elongation and a much more pronounced torsion of the entire metaphysis of the coracoid at the glenoid coracoid interface toward the antero-inferior plane, resulting in a deeply curved glenoid notch situated between the metaphyseal interface of the inferior two-thirds of the glenoid cavity and the glenoid-coracoid interface surface (Figure 8).

Integration of the metaphyseal surface of the coracoid in the Glenoid-Coracoid Interface and the Coracoid Process Epiphysis.

The results of the 2B-PLS analysis show that Axis 1, with a singular value of 2.303 × 10⁸, accounts for the entirety (100%) of the covariation between the two data blocks: the metaphyseal surface of the coracoid at the glenoid-coracoid interface and the coracoid process epiphysis. This finding indicates an exceptionally strong and well-defined relationship between these two structures, reflecting highly coordinated shape variation.

In contrast, the singular values for the remaining axes (Axis 2 and beyond) are negligible (< 0.001), suggesting that these axes capture only a minor fraction of the covariation. This pattern highlights that the first axis effectively encompasses almost all the shared variation, with subsequent axes likely representing noise or biologically insignificant variations.

Discussion

This study investigates the application of 3DGM to assess maturity changes in the coracoid process and its corresponding fusion site on the glenoid-coracoid interface. The shape transformations described by PC1 in both Shape and Form spaces provide a framework for understanding the maturation sequence of the coracoid process and its glenoid-coracoid fusion site.

Our findings reveal parallel morphological changes between the articular base of the coracoid process and its corresponding metaphysis within the glenoid-coracoid interface throughout skeletal development. Ogden and Phillips (1983) previously observed that the coracoid expansion around age 2 significantly influences the structural maturation of the glenoid-coracoid interface, fostering the establishment of a true bipolar growth interface. From the end of infancy onward, the coracoid epiphysis acquires a more complete morphology, which facilitates a detailed examination of its skeletal developmental trajectory. The developmental sequence in our study begins with the first maturity group, encompassing the complete childhood period from 3 to 7 years (Bogin, 2021). During this stage, the coracoid process is presented as short and slightly curved, with a rounded, flat articular base, while its associated metaphysis at the glenoid-coracoid interface shares a similarly short, flat, and rounded morphology. Throughout childhood, ossification changes in the scapula remain minimal, apart from the progressive replacement of epiphyseal cartilage along the vertebral border through advancing endochondral ossification (Rissech & Black, 2007). The only ossification event isolated from the scapular body during this period pertains to the continued and gradual development of the coracoid process (Kothary et al., 2014).

At the age of 7 years old, the transition from the first to the second maturity group marks a shift from childhood to preadolescence, coinciding with the onset of the midgrowth spurt (Bogin, 2021). This event represents a modest, temporary increase in growth velocity, generally observed around this age. However, sometimes the midgrowth spurt is difficult to detect. For instance, in the study by Dos Santos et al. (2019), this spurt was observed in approximately 70% of a Portuguese sample at an average age of 6.5 years, with minimal differences between sexes. In contrast, studies such as the First Zurich Longitudinal Study of Growth identified the midgrowth spurt between ages 6.5 and 7 in girls and between ages 7.5 and 8 in boys, indicating potential sex-related differences.

The midgrowth spurt has been linked to adrenarche, a developmental phase marked by increased androgen production from the adrenal glands, which are located above each kidney (Bogin, 2021). However, as noted by Butler et al. (1990) and Sheehy, Gasser, Molinari, and Largo (1999), caution is warranted when correlating the timing of the midgrowth spurt with adrenarche. For example, Remer and Manz (2001) suggest that a significant increase in androgen secretion may occur 1 to 2 years after the midgrowth spurt, based on a sample where this spurt appeared at an average age of 6.8 years.

In terms of skeletal development, the scapula’s first secondary ossification centre, the subcoracoid, typically emerges between ages 8 and 10, aligning with the timeframe of the second maturity group and coinciding with the previously mentioned events. The subcoracoid functions as a secondary centre for the coracoid itself, initially fusing with it and subsequently with the scapular body around age 16, thereby linking the coracoid process with the superior third of the glenoid cavity (Scheuer & Black, 2000).  The second maturity group, spanning from 7 to 10 years, is notably influenced by the emergence of the subcoracoid center. One of the most visible morphological changes in this phase is the pronounced extension of the subcoracoid articular area on the articular base of the coracoid process, which becomes increasingly distinct. This morphological transformation, coupled with an anteroposterior extension of the medial border of the coracoid base, causes the structure to lose the rounded, smoother appearance typical of the previous age group. This change is also mirrored in the morphology of the coracoid metaphysis within the glenoid-coracoid interface, which gradually acquires a more quadrangular form. Furthermore, the posterior side of the metaphysis in the interface begins to increase towards the superior plane. This marks the beginning of a transition in the plane of the coracoid metaphysis in the glenoid coracoid of the interface towards the infero-anterior direction. Around age 10, the onset of the third maturity group signals a transition from puberty into early adolescence, a shift also influenced by gonadarche, or the activation of the gonads (Bogin, 2021). This process triggers the active production of sex hormones, which are essential not only for the development of secondary sexual characteristics but also for the significant increase in growth rate that leads to peak height velocity (PHV) (Bogin, 2021; Falkner & Tanner, 1978; Hermanussen & Burmeister, 1993; Kumanov & Agarwal 2016; Phillip & Lazar, 2003; Rogol, 2010; Tanner, 1976). Therefore, the third maturity group, encompassing ages 10 to 14, is characterized by a marked acceleration in growth velocity in both height and weight during early adolescence. During this stage, the coracoid process not only elongates but also develops a more pronounced curvature, enhancing its hook-like form. Its articular base extends medially to laterally, with a noticeable expansion along the anterior-medial border. Thus, having an extension towards the metaphyseal surface of the coracoid angle. The metaphysis at the glenoid-coracoid interface reflects these morphological adjustments in the coracoid’s articular area, with the adding of an increasingly pronounced twist toward the infero-anterior plane.

In previous maturity groups, it has been established that the subchondral bone of the glenoid typically appears flat or slightly convex in an anteroposterior view. However, at the onset of the third maturity group, the subchondral bone of the glenoid gradually adopts a concave contour, becoming more similar to the articular surface. Additionally, around age 10, undulations begin to develop at the cartilage-bone interfaces of the acromion and the inferior scapular margin (Ogden and Phillips, 1983). These morphological changes are believed to be responses to alterations in the stress-strain dynamics from the surrounding musculature. In the coracoid process, a similar biomechanical response is suggested by the pronounced curvature of the coracoid body and the marked torsion observed in its metaphysis at the glenoid-coracoid interface. This morphological adaptation appears to reflect the increasing biomechanical demands and muscle mass growth characteristic of this stage of skeletal development, indicating the coracoid’s functional alignment with the evolving musculoskeletal and ligamentous system during early adolescence.

The fourth maturity group, ranging from 14 to 16 years, represents the final stage of skeletal maturity for the coracoid before its fusion. During this period, the coracoid process becomes more pronouncedly curved, and the area of its apex elongates. This transformation may be influenced by the appearance of two secondary ossification centres located at the apex and angle of the coracoid, which contribute to its structural maturation, though complete fusion is not achieved until approximately age 20. The articular surface of the subcoracoid substantially elongates as it spreads laterally, adapting to the curvature of the scapular notch’s morphology in the anterolateral region (). Whereas the metaphysis of the coracoid at the interface undergoes a pronounced torsion toward anterior, aligning finally within a sagittal plane as it approaches final fusion with the epiphysis.

This sagittal reorientation serves a clear biomechanical function that becomes more active from puberty onward, supporting the stability and mobility of the shoulder complex as the surrounding musculature develops (Rockwood & Matsen, 2017). The coracoid process is a critical anatomical site for the attachment of several key muscles and ligaments essential to shoulder mechanics. Key muscular attachments include the pectoralis minor, which links the coracoid process to the thoracic cage; the short head of the biceps brachii, which extends from the coracoid to the proximal radius; and the coracobrachialis, which connects the coracoid to the humerus. Furthermore, important stabilizing ligaments attach to the coracoid, including the coracoclavicular ligaments—the conoid and trapezoid—which are vital for maintaining clavicular alignment and stability. The coracoacromial ligament, one of the primary extracapsular ligaments of the glenohumeral joint, provides crucial support against superior humeral head displacement, while the coracohumeral ligament strengthens the joint capsule, enhancing overall structural integrity (Nordin & Frankel, 2001). Together, these muscular and ligamentous attachments underscore the functional and biomechanical importance of the coracoid process within the shoulder (Gasbarro et al., 2017).

The coracoid process undergoes significant morphological changes as it approaches skeletal maturity. Initially positioned more superiorly, the metaphysis of the coracoid gradually shifts through anterior torsion, ultimately aligning the coracoid epiphysis within a more sagittal plane (Figure 1b). During adolescence, rapid skeletal and muscular growth further accentuates this torsion, and the coracoid process adopts a distinct hook shape as its apex elongates (Figure 1b).

This period of skeletal maturation is further distinguished by the development of the remaining secondary ossification centres within the scapula. The rim of the glenoid cavity and the acromion appear first, typically between ages 14 and 16. Toward the later stages of adolescence, from approximately ages 15 to 17, additional ossification centres emerge along the medial border and the inferior angle of the scapula (Scheuer & Black, 2004, 2000). This progressive scapular development strengthens the shoulder girdle and glenohumeral joint, reinforcing the biomechanical adaptations seen in the coracoid process and supporting the increased demands on the shoulder complex during this critical growth period (Mao & Nah, 2004; Villemure & Stokes, 2009).

When comparing the Shape and Form space analyses, it is observed that while in the Shape Space Analysis of the coracoid process, all the groups generally overlap to the same degree, in the Form Space Analysis the last two groups overlap to a greater degree. This suggests that while the changes in shape and size are sufficiently critical in the first two maturity groups, this is not the case for the last two maturity groups (Figure 4 and 7). Furthermore, when comparing the degree of overlapping of the last two maturity groups between the Shape and Form space analyses, it is observed that the shape variant is more divisive than the size variant. Therefore, the changes in size are more significant in the transition from the first to the second maturity group, coinciding with the mid-growth spurt, and in the transition from the second to the third maturity group, coinciding with the adolescent growth spurt. While in the transition from the third to the fourth maturity groups, the shape variant is more discriminatory than the size variant. Similarly, in the case of the coracoid metaphysis at the glenoid-coracoid interface, both in the Shape and Form Space Analyses, the first two groups are also the most independent, especially in the shape analysis while the last two overlap, Therefore, the changes in shape and size of the first two maturity groups are more pronounced in contrast to the last two maturity groups (Figure 4 and 7).

Additionally, the results of the 2D-PLS analysis reveal a clear dominance of Axis 1 in explaining the total covariation, which underscores a functional integration and/or developmental coordination between the glenoid-coracoid interface and the coracoid process epiphysis. This shape covariation may be driven by the need to maintain shoulder joint stability or mobility. The lack of significant covariation in secondary axes supports the notion that the relationship between these structures is strongly concentrated in a single primary dimension of variation. This aligns with previous studies emphasizing modularity and integration in the evolution and development of the skeletal system

Conclusion

This study outlines a clear sequence of morphological transformations in the coracoid process and its metaphyseal interface as they progress through maturity stages, reflecting both structural development and adaptive functional changes. Comparative analysis of shape and form spaces shows that size variations are more influential in earlier growth stages, while shape differentiation plays a greater role in later stages than size, which highlights the value of geometric morphometrics in establishing shape differences independent of size variations. The observed influence of epiphyseal changes on the metaphysis supports the potential development of new methodologies for assessing skeletal maturity in dry bone contexts, especially valuable in anthropological and forensic settings where radiographic tools may be unavailable. This work provides a deeper understanding of developmental stages in dry bones, contributing to improved accuracy in skeletal age estimation.

Acknowledgements

We have benefitted from fruitful discussions with our colleagues from the Laboratorio de Evolución Humana at the University of Burgos

Grants

The Atapuerca research project is financed by MCIN/AEI/10.13039/501100011033/FEDER, UE grant number PID2021-122355NB-C31.

References

Bastir, M., García Martínez, D., Recheis, W., Barash, A., Coquerelle, M., Rios, L., Peña-Melián, Á., García Río, F., & O’Higgins, P. (2013). Differential growth and development of the upper and lower human thorax. PLOS ONE, 8(9), e75128. https://doi.org/10.1371/journal.pone.0075128

Bastir, M., & García Martínez, D. (2015). Asymmetry in the human thorax. Journal of Human Evolution, 78, 22-33. https://doi.org/10.1016/j.jhevol.2014.12.008

Bayley, N. (1946). Tables for predicting adult height from skeletal age and present height. The Journal of Pediatrics, 28(1), 49-64. https://doi.org/10.1016/s0022-3476(46)80086-6

Bogin B. (2021). Patterns of human growth. 3rd Edition Vol. 88. Cambridge: University Press. doi: 10.1017/9781108379977

Bookstein, Fred L. (2018). A Course in Morphometrics For Biologists: Geometry And Statistics For Studies Of Organismal Form.

Bookstein, F. L. (1991). Thin-plate splines and the atlas problem for biomedi-cal images. In Proceedings of the 12th international conference on information processing in medical imaging (pp. 326–342). Springer-Verlag.

Butler, G. E., McKie, M., & Ratcliffe, S. G. (1990). The cyclical nature of prepubertal growth. Annals of Human Biology, 17(3), 177-198. https://doi.org/10.1080/03014469000000952

Cameron, N. (2004). Measuring maturity. En R. C. Hauspie, N. Cameron, & L. Molinari (Eds.), Methods in human growth research (pp. 108–140). Cambridge University Press.

Cardoso, H. F. V. (2008). «Age Estimation of Adolescent and Young Adult Male and Female Skeletons II, Epiphyseal Union at The Upper Limb and Scapular Girdle in A Modern Portuguese Skeletal Sample». American Journal of Physical Anthropology 137(1):97-105. doi: 10.1002/ajpa.20850.

Cardoso, H. F. V., & Ríos, L. (2011). Age estimation from stages of epiphyseal union in the presacral vertebrae. American Journal of Physical Anthropology, 144(2), 238-247. https://doi.org/10.1002/ajpa.21394

Cardoso, H. F., Pereira, V., & Rios, L. (2014). Chronology of fusion of the primary and secondary ossification centers in the human sacrum and age estimation in child and adolescent skeletons. American Journal of Physical Anthropology, 153(2), 214-225. https://doi.org/10.1002/ajpa.22422

Conceição, E., & Cardoso, H. (2011). Environmental effects on skeletal versus dental development II: Further testing of a basic assumption in human osteological research. American Journal of Physical Anthropology, 144(3), 463-470. https://doi.org/10.1002/ajpa.21433

Coqueugniot, H., & Weaver, T. D. (2007). Brief communication: Infracranial maturation in the skeletal collection from Coimbra, Portugal: New aging standards for epiphyseal union. American Journal of Physical Anthropology, 134(3), 424-437. https://doi.org/10.1002/ajpa.20683

Cowgill, L. W., & Schmidt, N. (2006). The effects of aging and sex on the mechanics of human cranial growth. American Journal of Physical Anthropology, 129(3), 512-522. https://doi.org/10.1002/ajpa.20452

Cowgill, L. W., Warrener, A., Pontzer, H., & Ocobock, C. (2010). Waddling and toddling: The biomechanical effects of an immature gait. American Journal of Physical Anthropology, 143(1), 52-61. https://doi.org/10.1002/ajpa.21289

Demirjian, A., Goldstein, H., & Tanner, J. M. (1973). A new system of dental age assessment. PubMed, 45(2), 211-227. https://pubmed.ncbi.nlm.nih.gov/4714564

Dos Santos, M. A. M., Baxter‐Jones, A., Reyes, A. C., Freitas, D., Henrique, R. S., Chaves, R. N., Bustamante, A., Tani, G., Katzmarzyk, P. T., & Maia, J. (2019). Are there gross motor coordination spurts during mid‐childhood? American Journal of Human Biology, 31(4), e23251. https://doi.org/10.1002/ajhb.23251

Eveleth PB, Tanner JM. 1990. Worldwide variation in human growth. Cambridge:       Cambridge University Press.

Falkner, F., & Tanner, J. M. (Eds.). (1978). Human Growth: 2 Postnatal Growth. Springer US. https://doi.org/10.1007/978-1-4684-2622-9

Freidline, S. E., Gunz, P., Harvati, K., & Hublin, J. (2013). Evaluating developmental shape changes in Homo antecessor subadult facial morphology. Journal of Human Evolution/Journal of Human Evolution, 65(4), 404-423. https://doi.org/10.1016/j.jhevol.2013.07.012

Frontier, S. (1976). Étude de la décroissance des valeurs propres dans une analyse en composantes principales: Comparaison avec le modèle du bâton brisé. J. Exp. Mar. Biol. Ecol. 25, 67–75.

García-González, R., Carretero, J. M., Rodríguez, L., & Arsuaga, J. L. (2019). Two new methodological approaches for assessing skeletal maturity in archeological human remains based on the femoral distal epiphysis. Archaeological and Anthropological Sciences, 11(12), 6515-6536. https://doi.org/10.1007/s12520-019-00920-6

García-González, R., Rodríguez, L., Muñoz-Guarinos, J., Sánchez-Puente, Z., Fernández-   Viejo, M., Cirotto, N., Navarro-Pérez, A., García Barreiro, M., Salazar-Fernández, A., Quintino, Y., Adán-Álvarez, G.†, Carretero, J.M. (2024). Paleodemographic profiles of the populations buried in San Pablo Convent (Burgos, Spain). ENTEMU, XX, 2024

García-Martínez, D., Bastir, M., Gómez-Olivencia, A., Maureille, B., Golovanova, L., Doronichev, V., Akazawa, T., Kondo, O., Ishida, H., Gascho, D., Zollikofer, C. P. E., De León, M. P., & Heuzé, Y. (2020). Early development of the Neanderthal ribcage reveals a different body shape at birth compared to modern humans. Science Advances, 6(41), eabb4377. https://doi.org/10.1126/sciadv.abb4377

Gasbarro, G., Bondow, B., & Debski, R. (2017). Clinical anatomy and stabilizers of the glenohumeral joint. Annals of Joint, 2, 58-58. https://doi.org/10.21037/aoj.2017.10.03

Gómez-Olivencia, A., Barash, A., García-Martínez, D., Arlegi, M., Kramer, P., Bastir, M., & Been, E. (2018). 3D virtual reconstruction of the Kebara 2 Neandertal thorax. Nature Communications, 9(1), 4387. https://doi.org/10.1038/s41467-018-06803-z

Gunz, P., & Mitteroecker, P. (2013). SEMILANDMARKS: A method for quantifying curves and surfaces. Hystrix, the Italian Journal of Mammalogy, 24(1). https://doi.org/10.4404/hystrix-24.1-6292

Harvati, K., Roding, C., Bosman, A., Karakostis, F., Grun, R., Stringer, C., Karkanas, P., Thompson, N., Koutoulidis, Moulopoulos, L., Gorgoulis, V. G., & Kouloukoussa, M. (2019). Apidima Cave fossils provide earliest evidence of Homo sapiens in Eurasia. Nature, 571(7766), 500-504. https://doi.org/10.1038/s41586-019-1376-z

Harvati K., Stringer C., Folorunso A.  (2024). Comparative 3D Shape Analysis of the Iwo Eleru Mandible, Nigeria. PaleoAnthropology. 10.48738/2024.iss1.xxx.

Hermanussen, M., & Burmeister, J. (1993). Children do not grow continuously but in spurts. American Journal of Human Biology, 5(6), 615-622. https://doi.org/10.1002/ajhb.1310050604

Karakostis, F. A., & Harvati, K. (2021). New horizons in reconstructing past human behavior: Introducing the “Tübingen University Validated Entheses-based Reconstruction of Activity” method. Evolutionary anthropology, 30(3), 185–198. https://doi.org/10.1002/evan.21892

Klingenberg, C. P. (2011). MorphoJ: an integrated software package for geometric morphometrics. Molecular Ecology Resources, 11(2), 353-357. https://doi.org/10.1111/j.1755-0998.2010.02924.x

Klingenberg, C. P. (2016). Size, shape, and form: concepts of allometry in geometric morphometrics. Development, Genes and Evolution, 226(3), 113-137. https://doi.org/10.1007/s00427-016-0539-2

Kothary, S., Rosenberg, Z. S., Poncinelli, L. L., & Kwong, S. (2014). Skeletal development of the glenoid and glenoid–coracoid interface in the pediatric population: MRI features. Skeletal Radiology, 43(9), 1281-1288. https://doi.org/10.1007/s00256-014-1936-0

Kumanov, P., & Agarwal, A. (Eds.). (2016). Puberty: Physiology and Abnormalities. Springer International Publishing. https://doi.org/10.1007/978-3-319-32122-6

Liversidge, H., & Molleson, T. (2004). Variation in crown and root formation and eruption of human deciduous teeth. American Journal of Physical Anthropology, 123(2), 172-180. https://doi.org/10.1002/ajpa.10318

Liversidge, H. M., Chaillet, N., Mörnstad, H., Nyström, M., Rowlings, K., Taylor, J., & Willems, G. (2006). «Timing of Demirjian’s Tooth Formation Stages». Annals of Human Biology 33(4): 454-70. https://doi.org/10.1080/03014460600802387

López-Costas, O., Rissech, C., Trancho, G., & Turbón, D. (2012). Postnatal ontogenesis of the tibia: Implications for age and sex estimation. Forensic Science International, 214(1-3), 207.e1-207.e11. https://doi.org/10.1016/j.forsciint.2011.07.038

Mao, J. J., & Nah, H.-D. (2004). Growth and development: Hereditary and mechanical modulations. American Journal of Orthodontics and Dentofacial Orthopedics, 125(6), 676-689. https://doi.org/10.1016/j.ajodo.2003.08.024

Mitteroecker, P., Gunz, P., Bernhard, M., Schaefer, K., & Bookstein, F. L. (2004). Comparison of cranial ontogenetic trajectories among great apes and humans. Journal of Human Evolution/Journal of Human Evolution, 46(6), 679-698. https://doi.org/10.1016/j.jhevol.2004.03.006

Mitteroecker, P., & Gunz, P. (2009). Advances in geometric morphometrics. Evolutionary Biology, 36(2), 235-247. https://doi.org/10.1007/s11692-009-9055-x

Mitteroecker, P., & Gunz, P. (2011). The evolution of human growth and development. PLOS ONE, 6(6), e20731. https://doi.org/10.1371/journal.pone.0020731

Mitteroecker, P., Gunz, P., Windhager, S., & Schaefer, K. (2013). A brief review of shape, form, and allometry in geometric morphometrics, with applications to human facial morphology. Hystrix, the Italian Journal of Mammalogy, 24(1). https://doi.org/10.4404/hystrix-24.1-6369

Mitteroecker, P., & Schaefer, K. (2022). Thirty years of geometric morphometrics: Achievements, challenges, and the ongoing quest for biological meaningfulness. American Journal of Physical Anthropology, 178(S74), 181-210. https://doi.org/10.1002/ajpa.24531

Mori, T., & Harvati, K. (2019). Basicranial ontogeny comparison in Pan troglodytes and Homo sapiens and its use for developmental stage definition of KNM‐ER 42700. American Journal of Physical Anthropology, 170(4), 579-594. https://doi.org/10.1002/ajpa.23926

Morley, J., Bucchi, A., Lorenzo, C., & Püschel, T. A. (2022). Characterizing the body morphology of the first metacarpal in the Homininae using 3D geometric morphometrics. American Journal of Physical Anthropology, 177(4), 748-759. https://doi.org/10.1002/ajpa.24473

Nordin M, Frankel V. (2001). Basic biomechanics of the muscoskeletal system. 3rd ed. Lippincott Williams & Wilkins.

Phillip, M., & Lazar, L. (2003). The regulatory effect of hormones and growth factors on the pubertal growth spurt. The Endocrinologist, 13(6), 465-469. https://doi.org/10.1097/01.ten.0000098609.68863.ab

Remer, T., & Manz, F. (2016). The Midgrowth Spurt in Healthy Children Is Not Caused by Adrenarche.

Rissech, C., & Black, S. (2007). Scapular development from the neonatal period to skeletal maturity: A preliminary study. International Journal of Osteoarchaeology, 17(5), 451-464. https://doi.org/10.1002/oa.890

Rissech, C., Schaefer, M., & Malgosa, A. (2008). Development of the femur—Implications for age and sex determination. Forensic Science International, 180(1), 1-9. https://doi.org/10.1016/j.forsciint.2008.06.006

Rissech, C., López-Costas, O., & Turbón, D. (2013). Humeral development from neonatal period to skeletal maturity—application in age and sex assessment. International Journal of Legal Medicine, 127(1), 201-212. https://doi.org/10.1007/s00414-012-0713-7

Rissech, C., & López-Costas, O. (2016). Development of the humerus: Implications for forensic age and sex estimation. Forensic Science International, 258, 158-165. https://doi.org/10.1016/j.forsciint.2015.09.010

Rissech, C., & Turbón, D. (2017). A study of human skeletal maturity using 3D imaging. Forensic Science Review, 29(2), 214-225. https://doi.org/10.1016/j.forsciint.2017.02.002

Roche, A. F. (1992). Growth, maturation, and body composition: The Fels Longitudinal Study 1929-1991. Cambridge University Press.

Rockwood, C. A., Matsen, F. A., Wirth, M. A., Lippitt, S. B., Fehringer, E. V., & Sperling, J. W. (2016). Rockwood and Matsen’s the Shoulder. Elsevier. https://doi.org/10.1016/B978-0-323-29731-8.00023-4

Rohlf, F. J., & Slice, D. (1990). Extensions of the Procrustes Method for the Optimal Superimposition of Landmarks. Systematic Zoology, 39(1), 40. https://doi.org/10.2307/2992207

Rogol, A. D. (2010). Sex Steroids, Growth Hormone, Leptin and the Pubertal Growth Spurt. En S. Loche, M. Cappa, L. Ghizzoni, M. Maghnie, & M. O. Savage (Eds.), Endocrine Development (Vol. 17, pp. 77-85). S. Karger AG. https://doi.org/10.1159/000262530

Scheuer, L., & Black, S. M. (2000). Developmental juvenile osteology. Academic Press.

Scheuer, L., & Black, S. (2004). The Juvenile Skeleton. Elsevier Science.

Sheehy, A., Gasser, T., Molinari, L., & Largo, R. H. (1999). An analysis of variance of the pubertal and midgrowth spurts for length and width. Annals of Human Biology, 26(4), 309-331. https://doi.org/10.1080/030144699282642

Slice, D. (2006). Modern Morphometrics in Physical Anthropology. 10.1007/0-387-27614-9.

Stull, K. E., L’Abbé, E. N., & Ousley, S. D. (2014). Using multivariate adaptive regression splines to estimate subadult age from diaphyseal dimensions. American Journal of Physical Anthropology, 154(3), 376-386. https://doi.org/10.1002/ajpa.22522

Stevenson, P. (1924). «Age Order of Epiphyseal Union in Man». American Journal of Physical Anthropology 7(1):53-93. doi: 10.1002/ajpa.1330070115.

Stewart, T. D. (1934). «Sequence of Epiphyseal Union, Third Molar Eruption and Suture Closure in Eskimos and American Indians». American Journal of Physical Anthropology 19(3):433-52. doi: 10.1002/ajpa.1330190317.

Tanner, J. M., Whitehouse, R. H., Hughes, P. C. R., & B.S., C. (1976). Relative importance of growth hormone and sex steroids for the growth at puberty of trunk length, limb length, and muscle width in growth hormone-deficient children. The Journal of Pediatrics, 89(6), 1000-1008. https://doi.org/10.1016/S0022-3476(76)80620-8

Todd, T. W. (1930). The Anatomical Features of Epiphysial Union. Child Development, 1(3), 186. https://doi.org/10.2307/1125704

Ogden JA, Phillips SB. (1983). Radiology of postnatal skeletal development. VII. The scapula. Skeletal Radiol;9(3):157–169

Villemure, I., & Stokes, I. A. F. (2009). Growth plate mechanics and mechanobiology: A survey of present understanding. Journal of Biomechanics, 42(12), 1793-1803. https://doi.org/10.1016/j.jbiomech.2009.05.021

Zelditch, M. L. (2004). Geometric morphometrics for biologists: A primer. Elsevier Academic Press.

Ziegler, D., & Gunz, P. (2012). A new method for shape analysis of the pelvis and its application to human evolution. Journal of Human Evolution, 63(1), 57-70. https://doi.org/10.1016/j.jhevol.2012.04.001

Author contributions

Azahara Salazar-Fernández: Conceptualization (supporting); data curation (equal); formal analysis (equal); investigation (equal); methodology (equal); visualization (equal); writing – original draft (equal); writing – review and editing (equal).

José Miguel Carretero: Data curation (lead); investigation (equal); methodology (equal); visualization (equal); writing – review and editing (equal); funding acquisition- project administration (lead).

Laura Rodriguez: Data curation (equal); formal analysis (equal); investigation (equal); methodology (equal); visualization (equal); writing – review and editing (equal).

Rebeca García-González: Conceptualization (lead); data curation (equal); formal analysis (equal); investigation (equal); methodology (equal); visualization (equal); writing – original draft (equal); writing – review and editing (equal).

Received: November 14^th, 2024;

Accepted: February 6^th, 2025 ;

Online first: February 20^th, 2025;

Published: TBA

Copyright: © 2024 Salazar-Fernández et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.