Intra- and interspecific variability among congeneric Pagellus otoliths

Otolith features are useful tools for studying taxonomy, ecology, paleontology, and fish biology since they represent a permanent record of life history. Nevertheless, the functional morphology of otoliths remains an open research question that is useful to completely understand their eco-morphology. This study aims to deepen the knowledge of intra- and interspecific variation in sagitta morphology in three congeneric seabreams, to understand how such variability could be related to the lifestyles of each species. Therefore, the sagittae (n = 161) of 24 Pagellus bogaraveo, 24 Pagellus acarne, and 37 Pagellus erythrinus specimens, collected from the south Tyrrhenian Sea, were analyzed using scanning electron microscopy and a stereomicroscope to assess morphometric features, variability between otolith pairs and the external crystalline structure the of sulcus acusticus. Statistical analysis demonstrated that, between the species, variability in sagittal otolith rostral length growth and sulcus acusticus features, correlated with increased fish total length and body weight. Moreover, slight differences between otolith pairs were detected in P. acarne and P. erythrinus (P < 0.05). The results confirm changes in otolith morphometry and morphology between different congeneric species and populations of the same species from different habitats.

. Morphometric mean values with standard deviation (SD) and range of P. bogaraveo and P. erythrinus juvenile group individuals: OL (otolith length), OW (otolith width), OP (otolith perimeter), OS (otolith surface), SP (sulcus perimeter), SS (sulcus surface), SL (sulcus length), SW (sulcus width), CL (cauda length), CW (cauda width), OSL, (ostium length), OSW (ostial width), RW (rostrum width), RL (rostrum length), CI (circularity), RE (rectangularity), aspect ratio (OW/OL %), the ratio of otolith length to total fish length (OL/TL), percentage of the otolith surface occupied by the sulcus (SS/OS%), percentage of the sulcus length occupied by the cauda length (CL/SL%), percentage of the sulcus length occupied by the ostium length (OSL/ SL%), rostrum aspect ratio (RW/RL%) and percentage of the rostrum length occupied by the otolith length (RL/OL%). The morphometric data of Pagellus bogaraveo shown in the table relate only to the left otolith since no significant difference was found between the left (L) and right (R) sides. www.nature.com/scientificreports/ did not penetrate as deeply as in the other two seabreams species (Fig. 3d-f). Pagellus erythrinus individuals did not display significant differences between juveniles and adults, although significant differences in circularity (OP 2 /OS), rectangularity (OS/(OL × OW)), and sagitta aspect ratio (OW/OL %) were detected between right and left sagittae (Supplementary Table S1). In the juvenile group, a negative correlation between sagitta length to total fish length ratio (OL/TL), fish weight, and total length was highlighted. A statistically significant positive correlation was observed between rostrum aspect ratio (RW/RL %) and fish weight, sagitta aspect ratio (OW/OL%) and fish weight, and sagitta aspect ratio (OW/OL%) and fish total length (Supplementary Table S2).
The negative correlation between sagitta length to total fish length ratio (OL/TL), fish weight, and total length was also observed in the adult group, whereas a significant positive correlation was detected for relative sulcus area percentage (SS/OS %) and fish weight, and relative sulcus area percentage (SS/OS%) and fish total length (Supplementary Table S2).
The mean shape of otoliths differed significantly between P. bogaraveo, P. erythrinus, and P. acarne (P < 0.001), although minor differences were observed between P. bogaraveo and P. acarne. The otolith contours are shown in Fig. 1a. The first two axes (PC1 and PC2) of the PCA plot showed a separation of otoliths contours between the three fish species. In particular, PC1 (80%) separated P. erythrinus otolith shape from that of P. bogaraveo and P. acarne, which overlapped on the left side of the diagram along PC2 (20%) (Fig. 1b). Marked differences in the otoliths shape have also been confirmed by LDA. From the LDA plot of the first two discriminant functions, we can see that Pagellus species were quite well separated (Fig. 1c).

Scanning electron microscopy (SEM) analysis. Among the otoliths in all the examined species, SEM
showed clear changes in the shape, size, and direction of the external textural organization of the sulcus acusticus and differences in the surface of the crista superior and inferior between juvenile and adult individuals at the intra-specific level.
In juveniles, the crista superior and inferior sloped gently toward the sulcus acusticus depression, with an almost flat surface (Figs. 2a, 3a). The sulcus surface appeared smoother than in adults, with several tips distributed over the entire sulcal surface (Figs. 2b, c, 3c). In juveniles, the external textural organization of the sulcus was composed of smaller and thin crystals sometimes melted together or embedded in organic materials. In comparison, however, the crystals in adult specimens had become larger and thicker (Figs. 2b, f, 5a-h). The crystals of juveniles were grouped, with rounded edges, and slightly orientated in the vertical and oblique planes, with the long axis of crystals following the incremental growth direction of the otoliths (from the nucleus to the outer edge of the growth) ( Fig. 5a, b, e, f). Not all the crystals had the same shape and external 3D organization as in the adults; they had a smoother surface with a more compact structure.
Regarding adult specimens, the crista surface steeply declined towards the sulcus, with a hollow preceding the sulcal depression (Figs. 2d, 3d, 4a-d). The external sulcal structure was more complex with a more textured, rougher surface than in juveniles. The crystals were narrowed, with a prismatic shape. They had sharp edges, were almost equally sized, longer, and had a more chaotic orientation than in juvenile individuals (Fig. 5c, d, g, h).
Our results showed that the sulcus of Pagellus individuals from the south Tyrrhenian Sea was heterosulcoid. Furthermore, a greater size difference in the ostium and cauda was observed in adult P. erythrinus and P. bogaraveo compared to juveniles. Generally, the cauda was larger than the ostium and markedly different in shape. Despite the ostium, cauda growth during fish development indicated a most pronounced heterosulcoid character in adult individuals than in juveniles.
The macroscopic structure and shape of sulcus acusticus displayed in the SEM images exhibited inter-specific differences between the three congeneric seabreams. The ostium of P. acarne otoliths was deeper than that in the other congeneric species, and it displayed, with those of P. bogaraveo, according to with the ostium and cauda shape classification 1 , a funnel-like shape with concave ostium walls that expand and broaden anteriorly from the region of confluence with the cauda. Unlike the other seabreams examined in this study, the shape of the ostium in the P. erythrinus sulcus acusticus was more rectangular, with a markedly tubular shape, and the cauda was distinctly curved, especially in juvenile specimens.

Discussion
Intra-and interspecific differences: comparison with former studies on Pagellus species and other fish species. To understand the relationship between function, shape, and the environment, it is essential to include the morphological variability of otoliths, considering biological and environmental variability leads to otolith shape heterogeneity through morpho-functional adaptation to different habitats. Several authors have highlighted changes in otolith shape between species and, in many cases, among populations of the same species (e.g., herrings, salmonids, and lutjanids). The intra-specific variability of otolith morphology and shape are the basis of stock separation and assessment and is related, especially in sagittae, with environmental (e.g., water temperature, salinity, and depth) and biological factors (e.g., sex, ontogeny, and genetic variability) 20 .
The analysis of the three Pagellus species revealed that otolith morphology and morphometry did not follow those described in a previous study 1 conducted in the western Mediterranean Sea and the Atlantic Ocean in term of rectangularity, circularity, sagitta aspect ratio and sagitta length to total fish length ratio. Although the images provided in our study closely resembled those from research in other geographical areas, the morphometric measures (obtained according to the procedures and methods described in the previous literature 1,20,23,32 ) exhibited several differences. Considering the scale of our study compared to previous studies, it is difficult to provide an entirely valid comparison; the differences in sagitta morphology and morphometry could have been triggered by biotic and abiotic parameters (e.g., temperature, salinity, genotype, habitat type, differences in food quality and quantity) 13,[33][34][35] . Such environmental and genetic factors may be primary drivers of otolith morphometry and morphology among fishes in different habitats. Therefore, detected shape differences are at the basis of fish stock differentiation 36     Our results indicate that the min-max circularity and rectangularity of P. bogaraveo from the Southern Tyrrhenian Sea differ from those calculated in a previous study 1 in the western Mediterranean Sea, and the north and central-eastern Atlantic ocean. Moreover, the increase in circularity in larger specimens, confirms a greater tendency toward circular than elliptical otolith shape in southern Tyrrhenian Sea species compared to those in other Mediterranean and Atlantic areas.
Despite statistical differences and correlations in this study supported the hypothesis that some changes in sagitta morphology are related to fish size differences, several aspects and studies should be performed to better understand this relation. The negative correlation between the ratio of sulcus acusticus surface to the entire sagitta, rostral morphology, and the increase in specimen's size was related to the expansion in the length and surface of the entire sagitta and rostral area in larger specimens. These features, with no statistical relevant increment   , juveniles (a, b) and adults (c, d) www.nature.com/scientificreports/ in sulcus acusticus surface and increased rostrum length, could be correlated with more pronounced peripheral sagitta growth in this species. Sagitta, in fact, after fish pelagic phase, might increases its surface in the rostral area and the margins. Since the present study did not take into account ontogenetic stages and specimens age, it is hard to relate this result with sagitta and sulcus acusticus growth. But reading this increase by an ecological point of view, it could be related to the lifecycle of the species. During the juvenile stage, in the early stage of pelagic life, the species inhabits shallow water. Adults inhabit deep-water environments, migrating down the continental slope to a depth of 800 m after the juvenile stage. These changes in habitat might be the cause of morphological variations in the sagittae, highlighting the relationship between sagitta features and environmental and biological factors.
The P. acarne specimens demonstrated the highest number of morphometrical parameters that did not follow those of the same species described in a previous study (i.e., circularity, rectangularity, sagitta length to total fish length ratio, and sagitta aspect ratio) 1 . These morphometrical changes are reflected in otolith shape. The otoliths from specimens in our study were largely circular, with highly irregular margins and a rostrum that varied in length and width through the left and right sagitta, as indicated by the significant differences in rostrum aspect ratio values.
The morphometrical results in P. erythrinus revealed differences in circularity, rectangularity, sagitta length to total fish length ratio and sagitta aspect ratio compared to a previous study 1 in the western Mediterranean Sea and north-central eastern Atlantic ocean.
The P. erythrinus specimens were characterized by a pentagonal otoliths shape, and increased circularity compared to the same species from other areas. The results also indicated small differences between the left and right sagitta. This small differences were previously described in other Mediterranean sub-areas, for example, otolith width values in P. erythrinus specimens collected in the Gulf of Tunisia 37,38 .
As said above for P. bogaraveo, it is hard to relate the differences between juveniles and adults with fish growth due to the absence in present paper of ontogenetic and age analysis. The higher width than length, demonstrated by min-max width values in Tables 1 and 2, in sagittae of adults P. erythrinus specimens could be correlated with an exponential increment in fish size compared to the sagittal length. Further analyses on ontogenetic development of this species are required to better define the sagitta growth related to fish growth.
The increase in sulcus acusticus surface exhibited in the adult specimens could be correlated with feeding habits; during its adult life, this species is a benthic feeder and inhabits deeper environments than juveniles 28,29 .
Our study confirmed slight differences between width values in left and right sagitta previously described in P. erythrinus and extend the differences to other parameters, such as circularity and rectangularity (Tables 1,  2). Concerning P. acarne, however, marginal differences between the left and right sagittae were observed for the first time.
This slight differences are supported by the literature concerning genetic and environmental stressors 41 . Since the functional morphology of otoliths is not completely understood, it is difficult to find a direct link between these small differences and the ecology of the species. However, several eco-functional factors, such as feeding behavior, deserve attention as fundamental for a better understanding of the relationship between otolith features and species habitat. For example, P. erythrinus largely preys on strictly benthic organisms, such as polychaetes, brachyuran crabs, and benthic crustaceans. Most of these species frequently escape predators by hiding under the sandy substrate. Other Sparidae (Lythognathus mormyrus, Linnaeus, 1758) feed on benthic fauna, engulfing sediment and filtering it in the buccal cavity, demonstrated by the high percentage of detritus and benthic remains (e.g., scales, urchin spines, and benthic foraminifers) in the gut and stomach contents 29 . To engulf sediment, P. erythrinus performs a particular movement with the head and body, laterally shifting and pushing forward, to dig the bottom sand and reach prey. This kind of behavior, common in all benthopelagic species with the same feeding habits, could influence the sagitta growth and morphology, triggering small differences between the left and right sagitta. Further studies on this and other species with this behavior (e.g., L. mormyrus) are necessary to confirm this hypothesis.
Concerning inter-specific differences in sagitta morphology among the three species, it is difficult to read the results obtained in this study eco-morphologically since an insufficient understanding of the functional morphology and physiology of otoliths prohibits a direct relationship, valid for all the species, between eco-functional features and otolith morphology. Nevertheless, as expected, the shape analysis ( Fig. 1) revealed clear differences between the three congeneric species. Considering several ecological, functional, and biological features in each species, the results have demonstrated a sagitta morphology that could be in accordance with the ecology and lifestyle of these three congeneric seabreams.
Relationship between otolith morphology and ecology/lifestyle. The sagittae of P. acarne exhibited a shape resembling those in other pelagic species, with a long rostrum and the entire sagitta elongated and narrower than those in other two seabream species. The species that show the most pelagic habits, with largely planktivorous feeding at a small size, adapt also to benthopelagic feeding activity in adult life. The statistically relevant similarity found in P. bogaraveo could be proof of the ecomorphological adaptation of sagittae to pelagic and demersal environments. This hypothesis may be confirmed by marked differences in shape compared to those in P. erythrinus, which is the most benthic among the three species. www.nature.com/scientificreports/ Pagellus erythrinus was the species with the shortest rostrum. It also has the most benthic habits, largely preying on epibenthic and infaunal species. Moreover, its ecology and life cycle differ among the three species under study since they are strictly related to the benthic environment. This lifestyle could be in accordance with the differences observed in the shape analysis results. The sagitta contours appeared more circular and wider than those in the other two species. The PCA and LDA also confirmed the most difference in shape among the three species.
The species with the most marked antirostrum and sagitta shape was P. bogaraveo, which is a cross between the other two congeneric species. Pagellus bogaraveo is a demersal species, which inhabits the deep biocenosis and feeds in both benthic and mesopelagic environments. Furthermore, the ecology of this species could support the sagitta shape described in our study 27,30 .
Otolith morphology and morphometry in congeneric Pagellus species described in this study has followed the relationship between sagittal parameters, habitat, and depth described in previous literature 15 . According to several authors, the percentage of species with large otoliths increases with depth, except for abyssal depth. The specimens of P. bogaraveo analyzed in this paper (especially adult individuals) had larger otoliths than the other two Pagellus species due to their demersal habits (they inhabit the continental slope to a depth of 800 m). A larger sagitta is essential in demersal environments to compensate for light reduction by providing improved acoustic communication, sound perception 15,45 , and a sense of equilibrium 46 .

Sulcus shape. Considering the sulcus acusticus, in the otolith atlas for the western Mediterranean Sea and
Atlantic ocean 1 , studies describing and comparing otoliths 10 and the diversity and variability of otoliths in teleost fishes 9 , the sulcus in P. bogaraveo, P. acarne and P. erythrinus was described as heterosulcoid, with an ostium shorter than the cauda and a long, narrowed rostrum, especially in adult P. bogaraveo and P. acarne individuals. Heterosulcoid otoliths were also observed in south Tyrrhenian Sea Pagellus individuals, with marked differences between juvenile and adult specimens. In a demersal species, such as P. bogaraveo, juveniles live in shallow, coastal water. Once adults, they inhabit deeper water (to a depth of 800 m). Changes in the crystalline and morphological structure of sulcus acusticus between juveniles and adults reflect this species' need to adapt to deeper environments with less light.
The results indicate that in P. bogaraveo, the sulcus acusticus does not differ in surface between juvenile and adult specimens. This feature could be correlated with earlier sulcus acusticus development in this species, compared to P. erythrinus and P. acarne, emphasizing the role of the sulcus acusticus in this demersal species 48,49 . This might also confirm the strict correlation between biological and environmental factors and sagitta morphology in studied seabreams species.
Another morphological feature of the sulcus acusticus, which might support the ecology of the species, is the deep ostium and cauda. In adult specimens of P. bogaraveo and P. acarne, the sulcus structure deeply penetrated in the sagitta carbonate structure. Conversely, in adult P. erythrinus specimens, the sulcus did not penetrate as deeply as in the other two seabream species. This sulcal feature could correspond with the ecology and feeding behavior of P. erythrinus, which specializes in benthic strategies, including small differences between left and right sagitta and the absence of the notch and antirostrum in sagittae.
Although the deeper sulcus acusticus in P. bogaraveo and P. acarne might be linked to depth distribution, as in P. bogaraveo, it may also correspond with high mobility related to feeding behavior, as in both P. bogaraveo and P. acarne. The different depths of sulcus acusticus can change the thickness of the otolithic membrane, by varying the relative motion of otoliths with the macula sacculi 2 . As previously demonstrated 50 , the different thicknesses of the otolithic membrane induce differences in mechanical resistance between the otolith and sensory epithelium.
The differences in sulcus acusticus and otolith ratio between P. bogaraveo specimens and the other congeneric species, demonstrated by the results, might be also correlated to the differences in habitat, feeding habits, and soundscape.
Despite the lack of information concerning the physiological ear response related to variations in macula or sulcus size, the sensory hair cells in macula sacculi are likely to be affected by changes in sulcus depth, shape, 3D structure (planar vs. curved), and surface. The significant difference in relative sulcus area may be due to typical alteration in this parameter concerning differences in the mobility patterns, food, feeding behavior, and spatial niche.
Higher relative sulcus area ratios have been observed in the deepest species or those with high mobility 49 . In our study, the morphometry results concerning the sulcus did not follow those in the previous literature, displaying higher values in P. erythrinus and P. acarne compared to P. bogaraveo, although the latter inhabits a deeper environment than the other congeneric species.
This higher relative sulcus acusticus surface and the larger, curved sulcus acusticus of P. acarne and P. erythrinus could be correlated with higher mobility in these species (especially P. acarne). In P. erythrinus, however, these features might be related to its benthic lifestyle.
As demonstrated by the PCA and LDA of sulcus acusticus parameters, P. erythrinus and P. acarne, which share similar depths and habitats, revealed marked similarities, whereas P. bogaraveo, which lives in the deepest strata of the water column, displayed the most different sulcus acusticus. However, PCA and LDA indicated that the otolith shape in the entire P. erythrinus sagitta was significantly different compared to those in P. acarne and P. bogaraveo.
These features could provide a reading key for sagitta and sulcus acusticus eco-morphology in the life cycle and environmental adaptation of fish.
The connection between the otoliths and the macula sacculi is fundamental for transducing environmental acoustic signals and for the relative motion of fish (balance). The sulcus acusticus is the area of the otoliths in which this connection occurs. www.nature.com/scientificreports/ Features of the texture. Furthermore, the external textural organization 23 changes between juveniles and adults or when environmental changes occur. The differences in the external textural organization found in juveniles and adults support those reported in the literature concerning other species 39 . Figures 3b, c and 5a-h, demonstrate that our study supported this prediction. However, P. bogaraveo and P. erythrinus juveniles, compared with other species (such as gurnards) 23 displayed a more uniform, mineralized, external textural organization. According to previous literature 8 , improved hearing capabilities in a species are closely related to a higher value of relative sulcus area ratio. Habitat features, such as depth, feeding strategies, mobility, trophic distribution, and ontogeny, could also influence this ratio.
Hence, it may be concluded that morphological differences in sulcus acusticus shape and surface among species are important for comprehending the ecomorphological and eco-functional role of sagitta 2,48 .
Comparing the intra-specific differences indicated by our results with those in the literature, discussing other populations, we cannot determine whether site-differences observed in sagitta shape are related to genetic evolution and/or adaptative response to environment. To make this distinction it would require a specific experiment in which offspring from different populations are raised in a controlled environment.
Furthermore, the knowledge about physiology and functional morphology is insufficient to provide a clear correlation between inter-specific differences among the three congeneric Pagellus species and their ecological and functional features. However, differences in sagitta morphology and morphometry among these three Pagellus species may be related to differences in lifestyle, ecology, and biology since they follow the ecomorphological features of sagittae and species ecology described in the literature.

Concluding remarks
This study has considered a wide range of morphometric and morphological characteristics in Pagellus species otoliths. Despite excellent and detailed photographs provided in previous studies 9, 10 , this paper provides, to our best knowledge, the first shape analyses, using R software of P. bogaraveo and P. acarne otoliths, and the first accurate SEM analyses of P. bogaraveo, P. acarne, and P. erythrinus from the study area and other regions.
An overall image of P. bogaraveo sagitta and its morphometrical features was created. Due to SEM imaging, we obtained, for the first time, the most accurate image of otoliths in these species and their external textural organization. This preliminary study provides grounding for an improved understanding of the structure and eco-morphological role of the sagitta in the life cycle of this species. Other methodologies (e.g., X-ray diffraction, auditory sensitive measurement, CT scan) are needed to deeply investigate the physiology of the sagitta and its ecological adaptation to the environment. The results could aid stock identification and improve understanding of the distribution of different Mediterranean populations and their differences. Improved understanding of the phenotypic plasticity and ecomorphological role of otoliths could also serve to compare the structure, morphometry, and crystalline composition of otoliths in congeneric species of Pagellus from different populations, to evaluate how the sagittal structures and features change according to different environments and habitats. This approach is essential to evaluate how the morphometry and shape of different sagittal areas, such as the sulcus acusticus, change under different environmental pressures.
It is essential to deepen the knowledge about the later asymmetry between sagitta pairs in P. erythrinus and P. acarne since this may affect stock differentiation based on shape analysis between populations from different sub-areas. This feature of otolith morphometry and shape could be another response to environmental pressure, which may clarify the role of phenotypic plasticity in sagitta development.

Materials and methods
A total of 44 P. bogaraveo otoliths (n = 24), 46 P. acarne otoliths (n = 24), and 71 P. erythrinus otoliths (n = 37) were collected from trawled specimens in the southern Tyrrhenian Sea (GSA10) between March and October 2019. Fish specimen collection was authorized by the CAMP.BIOL.19 project 52,53 . Fish otoliths were collected as part of annual research surveys, all involving lethal sampling. No experiments were conducted, nor were surgical procedures performed. No procedures caused lasting harm to sentient fish, nor were sentient fish subjected to chemical agents. The care and use of collected animals complied with animal welfare guidelines, laws, and regulations set by the Italian Government.
Before otolith extraction, each specimen was measured (TL to the nearest mm), weighed (body weight (BW) to the nearest g), and dissected to evaluate the sex and the maturity stage, according to the codes of sexual maturity in fish (MEDITS, freely available at http:// archi mer. ifrem er. fr/ doc/ 00002/ 11321/). For accurate morphometric analysis and statistical comparison of the data, specimens of each species were divided into two groups, according to the sexual maturity codes (i.e., juvenile and adult individuals).
The sagittae were removed from the otic capsule and cleaned of tissue using 3% H 2 O 2 for 15 min, followed by Milli-Q water. The dry otoliths were stored inside an Eppendorf microtube.
A Leica M205C stereomicroscope with a built-in LEICA IC80 digital camera was used to collect digital images of the otolith samples (Supplementary Figures S2, S3, S4).
Each sagitta was photographed twice, once with the sulcus acusticus facing upwards and once with the annuli side facing up.
Before being converted into binary format for contour extraction using ImageJ 1.48p software 53 (freely available at http:// rsb. info. nih. gov/ ij/), the longest axis was used to orientate the images horizontally and vertically to capture clear sulcus acusticus images, according to the literature 20 . Morphometry. According  , aspect ratio (OW/OL%), the ratio of otolith length to the total fish length (OL/TL), the percentage of the otolith surface occupied by the sulcus (SS/OS%), the percentage of the sulcus length occupied by the cauda length (CL/SL%), the percentage of the sulcus length occupied by the ostium length (OSL/SL%), the rostrum aspect ratio (RW/RL%) and percentage of the otolith length occupied by rostrum length (RL/OL%).
Supplementary Figure S5 provides a schematic diagram of the measured features.
Otolith shape analysis. Analysis of otolith shape was performed using shape R, an open-source software package that runs on the R platform (R version 4.0.5). This package was specifically designed to study otolith shape variation among fish populations or species 54 . First, the stereoscope captured otolith images were binarized using a threshold pixel value of 0.05 (intensity threshold). Once the outline of each image was detected, the master file containing the analyzed specimen information (e.g., fish length, weight, origin, maturity, and sex) was linked to each extracted outline. Wavelet and Fourier coefficients, required for statistical analysis, were extracted and adjusted for allometric relationships with the fish lengths. The wavelet coefficient was also used to www.nature.com/scientificreports/ obtain the graph shown in Fig. 1a, which compares the mean otolith shapes of the analyzed species. The quality of the wavelet and Fourier reconstruction was estimated by comparing how it deviated from the otolith outline (Fig. 6a). The maximum number of Fourier harmonics to be displayed was set at 15. Finally, the graph presented in Fig. 6b was produced by running a specific function of the g-plots R package, to investigate how the variation in wavelet coefficients is dependent on the position along the outline.

SEM analysis.
A total of four P. bogaraveo otoliths, three P. acarne otoliths, and four P. erythrinus otoliths underwent SEM analysis as described in previous studies 55 . The samples were fixed in 70% alcohol for 48 h and subsequently dehydrated using a graded series of alcohol from 70 to 100% (1 h in each solution). To avoid the critical drying point, samples were placed on a stub (SEM-PT-F-12) using conductive adhesive tables (G3347) and left for 12 h at 28 °C. Finally, the samples were sputter-coated with 20 nm gold-palladium. The samples were examined using a Zeiss EVO MA10 operating at an acceleration voltage of 20 kV.
Data analysis. All statistical analyses were conducted using the following software: Prism V.8.2.1 (Graphpad Software Ltd., La Jolla, CA 92037, USA), R vegan package V.2.5, and PAST V. 2.7 56 . Selected morphological parameters (OP 2 /OS, OS/[OL × OW], OL/TL, OW/OL%, SS/OS%, RW/RL%, and RL/OL%) were analysed using an unpaired t test to highlight any significant differences between the right and left sides of the otoliths and between juvenile and adult specimens. Differences in morphological parameters between specimens of different species (at the same maturity stage) were analyzed using a one-way analysis of variance (one-way ANOVA). Additionally, sulcus acusticus parameters were subjected to a principal component analysis (PCA) based on a variance-covariance matrix and Linear Discriminant Analysis (LDA) to show differences between all the analyzed species.
Finally, the correlation between the measured parameters and fish weight and total length was tested using the Pearson correlation coefficient.
To determine differences in otolith contours, wavelet coefficients were used to analyze shape variation among species using an ANOVA-like permutation test. Moreover, shape coefficients were subjected to a PCA, based on a variance-covariance matrix, and LDA to obtain an overview of the differences in otolith shape between the congeneric species examined. The significance level was set at P < 0.05.