Patterns of Directional and Fluctuating Asymmetry in Southern Ocean Sea Urchins

Abstract

Bilateral symmetry is a fundamental organizational trait in many metazoans. However, deviations from this symmetry, manifested as directional or fluctuating asymmetries, offer valuable insights into developmental, functional, and environmental processes. This study quantified and characterized bilateral asymmetry in three related species of the genus Abatus using a dual approach that integrates linear and geometric morphometrics. Our analyses reveal consistent patterns of directional asymmetry across different species, with specific trends showing that the left body sides tend to be larger. In contrast, fluctuating asymmetry exhibited an inverse relation with directional asymmetry. Also, linear morphometric analysis showed no significant correlation between directional asymmetry and either sex or body size, while geometric analyses only identified subtle shape deviations related to size but not to sex. These findings allow us to discuss the possible origins of this trait, mainly related to developmental constraints due to reproduction or growth, or carried over on a genetic basis. Our results highlight the importance of combining different morphometric approaches to clarify complex patterns of morphological variation and emphasize the significance of asymmetry analyses in understanding evolutionary and ecological processes in irregular echinoids.

1. Introduction

Morphological symmetry refers to the repetition of parts or structures in different positions or orientations within an organism [1], forming the body planes observed in living beings. In metazoans, various types of symmetries have been described [2,3,4], which differ in the number of times a structure is repeated as well as in the types of geometric transformations to which the structure or part is subjected (e.g., rotation, translation, reflection) [1]. Asymmetry is the most basic type of symmetry, where organisms lack any axis, plane, or point of symmetry, meaning they exhibit irregular shapes [4]. However, some of these organisms may display polarity in their structures or bodies, which can give rise to differentiation in morphological patterns [5]. It is important to note that this “asymmetry” refers to the strict sense of the concept, i.e., a lack of symmetry, where no established body planes exist [6]. This type differs from radial or bilateral asymmetry, which involves deviations from such symmetries.

Radial symmetry involves the repetition and rotation of structures n times around a point or axis, generating n planes of symmetry. Hexactinellid sponges, cnidarians, and regular sea urchins are examples of this type of symmetry [7]. Other less common types of symmetry include biradial symmetry in some algae and flowers, spiral symmetry in snail or nautilus shells, and spherical symmetry in certain unicellular eukaryotes or in some demosponges, among others [8]. All of these, except for asymmetry, are known as complex symmetries and differ from the most common existing symmetry: bilateral symmetry [9]. Organisms with bilateral symmetry are characterized by having two mirrored body sides or traits, that is, the left side or trait is a mirror image of the right side or trait [4]. Since both sides typically share the same genetics and are exposed to very similar environmental conditions, it is expected that both sides will exhibit the same phenotype [10]. However, deviations from symmetry are common in organisms and are referred to as asymmetries. In organisms with a bilaterally organized body plan (bilateral symmetry), there are three types of asymmetries, which differ according to the distribution of the trait (difference between left and right sides) in the population.

The first type is known as fluctuating asymmetry, and it is defined as small deviations between the left and right sides of individuals, which lead to greater dispersion or deviation of the trait from a mean of zero (Figure 1A). This asymmetry results from a balance between minor random disturbances during development (known as developmental instability or noise) [11,12] and the regulatory mechanisms of organisms that ensure developmental stability [13]. However, not all the processes are random, and these small perturbations can be influenced by environmental factors (e.g., pollution) [14,15,16,17] or by genetic processes (e.g., hybridization) [18,19]. The above, combined with the fact that this type of asymmetry shows little to no heritability [13] and has been linked to biological fitness [20,21], allows it to be used as an individual response to developmental stress [22]. In contrast, antisymmetry is characterized by a bimodal distribution of the trait with a mean at zero (Figure 1B), which is biologically represented as a trait that can be present on either the left or right side of organisms. Since the trait shows a similar frequency on both sides, at the population level there is a mix of right- and left-sided individuals (e.g., the larger claw in fiddler crabs) [23].

Finally, directional asymmetry is characterized by a trait with a normal (unimodal) distribution, with a mean different from zero (Figure 1C), meaning the trait tends to occur more frequently on one side of the organism than the other within the population (i.e., most individuals are right-handed or left-handed) [24,25]. Some clear examples of this trend are human internal organs. Also, it is widely accepted that directional asymmetry is genetically determined in organisms [26,27]. Although this pattern also occurs in antisymmetry, in that case the direction of the trait is generally not heritable (i.e., the presence of the trait is genetically determined, but its direction tends to be random) [28]. Furthermore, the origin of these types of asymmetries can be varied, and some directional or antisymmetrical traits may have adaptive, functional, or developmental constraint-related implications [13,29,30]. These multiple origins mean that these deviations may vary intraspecifically, for example, during the development of organisms due to changes in morphological patterns, maturation, life stages differentiations, and trait expression [31,32,33], or vary between sexes due to internal structural processes related to the development, change, and maturation of reproductive organs or gonads [34,35,36]. Therefore, the influence of both types of asymmetries on biological fitness is variable. Also, it is important to note that, for a specific trait, fluctuating asymmetry is not mutually exclusive with the other two types of asymmetry, so in a population, directional asymmetry or antisymmetry can occur alongside significant fluctuating asymmetry.

The analysis of asymmetry has been a central focus in the study of shape and its variation in bilaterally symmetrical organisms [37,38]. Over time, two main approaches have stood out in analyzing asymmetry patterns that differ primarily in their focus and the type of shape measurement used. Since the early days of morphology, linear morphometrics (LM) has been a widely used method, particularly in systematics, taxonomy, and the description of new species [39,40,41,42]. To assess bilateral asymmetry, linear morphometrics uses one or several linear measurements of traits on both sides of the organism, which are then compared [43,44]. However, one of the challenges faced by this type of approach is that the shape of an organism is difficult to capture using linear measurements; further, many of these measurements may vary or not allometrically, making it difficult to compare between organisms or traits adequately [45]. Moreover, analyzing shape through their parts (linear measurements) does not allow for the assessment of the overall shape of organisms. In contrast, geometric morphometrics (GM) uses a geometric definition of shape, i.e., shape encompasses all the geometric features of an object except for its size, position, and orientation [46], and it is based on the use of homologous points among individuals or specimens located on key morphological structures (landmarks). Landmarks allow for a complete shape characterization through their spatial relation, and a differentiation between shapes through the deviation between them and between individuals [1]. Although geometric morphometrics is a useful tool for the analysis of asymmetry, it only allows us to assess whether bilateral asymmetry is present, how much it contributes to shape variation, and whether it is statistically significant. However, it does not reveal the specific pattern of deviation, such as side preference. On the other hand, linear morphometrics can detect and show these patterns. Therefore, a combined analysis using both approaches can provide a more detailed understanding of morphological patterns and a complete characterization of them.

In sea urchins, class Echinoidea, systematics has been primarily based on the evaluation of morphological and morphometric characters [47]. Therefore, symmetry (radial or bilateral) has played a crucial role in defining certain taxonomic clades. Echinacea and a few smaller groups are commonly known as regular sea urchins and represent slightly more than half of all extant echinoid species [48]. In these sea urchins, metamorphosis involves a transformation from bilaterally symmetrical larvae to adults with pentameral symmetry, reflecting their inclusion within the clade Bilateria. On the other hand, the remaining clade, known as irregular sea urchins or Irregularia, is characterized by the loss of pentameral symmetry and the development of a new secondary bilateral symmetry during ontogeny, caused by the migration of the anus from the apical to the posterior position [49,50]. The morphological and behavioral evolution of irregular sea urchins has led to major differences from their radially symmetrical ancestors, including obligate anterior locomotion, an infaunal lifestyle, mandibular modifications, and dietary changes [51]. Within the clade Irregularia, Spatangoida is currently the most diverse group of extant sea urchins [52]. Their bilateral symmetry is particularly visible and prominent compared to other orders within Irregularia, such as Cassiduloida or Clypeasteroida [49,53], and their distinctive test shape has even earned them the common name of heart urchins. Within this group, numerous deviations from bilateral symmetry have been documented, both in extinct and living species [49,54,55,56].

In the genus Abatus (Order Spatangoida; Family Schizasteridae), unlike other spatangoids, bilateral asymmetry has been studied only rarely. Stige et al. [57] measured both fluctuating and directional asymmetry in the oral plates of Abatus cordatus, finding significant values for both types. In their study, fluctuating asymmetry varied between the populations studied, while directional asymmetry was influenced by both the population and the body size of individuals. Lane [58], studying Abatus nimrodi in two areas near Casey Station in Antarctica that differ in contamination levels and types, observed irregularities in bilateral symmetry among individuals from more polluted sites, although no statistical methods were applied and no directional pattern was identified. Finally, Gil et al. [59] found asymmetries in the size of the brood pouches in Abatus cavernosus, with brood pouch IV (anterior left) being the largest and housing the highest number of offspring, a pattern thought to be related to the position of the three gonopores present in these species. Excluding this last study, no other work has shown a clear tendency (left, right, or both) in the observed asymmetries, only reporting variation between the two sides. In addition, apparent asymmetries also have been observed in some species, which seems to be systematic between individuals and species. Since males in Abatus lack brood pouches (sexual dimorphism), and these structures are developed in females at reproductive maturity (at a certain size/age) for brooding young [41,60,61], the possible asymmetry in Abatus could be related to sexual structures, with a larger body side potentially allowing higher brood pouch development or, conversely, the expansion of the pouch or other internal reproductive structures (in males or females) exerting pressure on the test and causing expansion of that side during growth.

Based on the apparent asymmetry in Abatus, this work aimed to assess the deviations of symmetry in these sea urchins and their relationship with biological features such as sex and size (age). This was achieved by characterizing and quantifying the bilateral asymmetry in three species phylogenetically, morphologically, and ecologically, which constitute a clade within Abatus [41,62,63,64], using two approaches: linear morphometrics and geometric morphometrics.

2. Materials and Methods

2.1. Sample Collection and Processing

To characterize and quantify bilateral asymmetry in Abatus, 61 individuals belonging to three species were obtained from three localities (Figure 2) [65]. Abatus agassizii specimens (n = 21; 11 males and 10 females) were collected from Fildes Bay (62.2199° S, 58.9533° W; King George Island) in Antarctica. Abatus cavernosus specimens (n = 20; 10 males and 10 females) were collected from Possession Bay (52.3353° S, 69.4782° W; Magellan Strait) in Patagonia. And A. cordatus specimens (n = 20; 10 males and 10 females) were collected from Port aux Français (49.3510° S, 70.2107° E; Kerguelen Islands). Previous studies have documented significant shape differences at the intraspecific level between A. cavernosus populations from Patagonia and the Falkland/Malvinas Islands, possibly due to phenotypic plasticity or divergence in evolutionary trajectories [62]. These phenotypic differences may reflect another pattern of asymmetry. Therefore, to assess this morphologically distinct population, an additional 11 individuals (11 males) of A. cavernosus from the Canache area (51.6963° S, 57.7851° W; Falkland/Malvinas Islands) in the Subantarctic province were included as a fourth locality, generating a total of 72 individuals for the analyses.

In addition, to evaluate only variations in asymmetry in relation to a wider size range, 81 individuals of Abatus cavernosus were obtained from the Possession Bay site (Figure 2). These individuals span a size/age gradient, from juveniles less than one year old to adults up to five years of age, which corresponds to the maximum size for these species [60]. All samples used in the morphological analyses were obtained during previous field campaigns and were sourced from the biological collection of the Millennium Institute for Biodiversity of Antarctic and Subantarctic Ecosystems (BASE). The specimens were preserved in 95% ethanol [65].

Sample preparation and cleaning followed the methodology used by Stige et al. [57]. Each individual was submerged in water to remove excess ethanol and then placed in a 5% bleach solution for two minutes to dissolve organic structures. Using a toothbrush, external structures of each sea urchin (e.g., spines, podia, and ambulacral feet) were removed to expose the test. Individuals were photographed in aboral (dorsal) view with a Nikon D7500 digital camera (18–55 mm lens). For morphometric analysis, the landmark method was used, based on homologous points across individuals that allow for the measurement of shape deviations (Klingenberg 2015) [1]. Based on the dimensionality formula for 2D morphometrics, “2k − 4 < n”, where k is the number of landmarks and n is the number of samples [66], 37 landmarks were estimated as the maximum number to capture the shape geometry without overfitting the analysis [62]. In each photograph (tps file), a curve of 37 equidistant points was digitized along the outline of the individual in a counterclockwise direction, starting and ending at ambulacrum III (following Loven’s system), using the software tpsDig2 version 2.31 [67]. The curve was then converted to landmarks using tpsUtil32 version 1.78 [67].

To estimate measurement error and the level of fluctuating asymmetry across species, landmark digitization was performed twice per individual at different times. Finally, to standardize the data by removing size, rotation, and translation effects among individuals and to extract shape information from the landmarks, a Procrustes analysis was performed [68].

2.2. Morphometrics Analyses

To characterize and quantify directional asymmetry in Abatus species, two morphometric methods were used: (1) linear morphometrics and (2) geometric morphometrics. To quantify fluctuating asymmetry in individuals, only geometric morphometrics was applied.

For the analyses using linear morphometrics, an irregular polygon was made for each side of the individuals by connecting all consecutive landmarks from 1 to 19 for the left side and from the landmark 37 to 19 for the right side (Figure 3A; hatched areas); the area of each polygon was extracted/quantified (in arbitrary units) using the Shoelace theorem [69]. With the area obtained from each side per individual, the asymmetry of the trait (BA) was calculated for each sea urchin in the datasets using formula (1).

BA = R − L

(1)

where “R” is the value of the right trait (right side area) and “L” is the value of the left trait (left side area), with BA representing the difference in size between sides (a higher BA indicates a higher level of asymmetry). The sign of the BA also indicates which side is larger: a positive BA indicates a larger right side, while a negative BA indicates a larger left side. To visualize the asymmetry curve at the genus level, a frequency histogram of individual BA values was generated. To qualitatively visualize the direction of asymmetry (left or right) for each species, species-specific bar graphs were produced. Subsequently, to determine whether the differences between sides (asymmetry trait) were significantly different from what would be expected by chance, statistical tests comparing group means at both the genus and species levels were performed. First, the normality of the data was assessed using the Shapiro–Wilk test [70], which showed different results depending on the variable. Based on these findings, and to avoid potential errors, a Mann–Whitney U test (non-parametric test) [71,72] was used. In addition, since both parametric and non-parametric statistical tests may have low power when sample sizes are small, two-tailed permutation tests were performed to compare the difference between sides and strengthen the asymmetry analyses [73].

Finally, the relation between directional asymmetry and individual biological features such as age and sex was statistically analyzed. To evaluate whether trait asymmetry differs between males and females, the normality of BA intensity was tested for each sex within the sample of 72 individuals using the Shapiro–Wilk test [70], revealing a non-normal distribution in both groups (p < 0.05). Consequently, a non-parametric test for differences in means between sexes was applied using the Mann–Whitney U test [71,72]. To assess whether the intensity of asymmetry varies with individual size, normality of BA values was tested in the two datasets: the 72 individuals from three species and the 81 individuals of A. cavernosus from Possession Bay (size/age gradient). The Shapiro–Wilk tests did not reveal evidence of a normal distribution in any of the size datasets (p < 0.05). Based on this, a Spearman correlation test (non-parametric) [74] was conducted for both datasets between the absolute value of asymmetry (|BA|) and the Csize variable (centroid size), obtained from the Procrustes analysis of the samples. The Csize variable is homologous to the linear body size (anterior-posterior length), showing an almost perfect and significant correlation between the two (Rho = 0.978; p < 0.0001). In addition to the mean difference and correlation tests, permutation tests were performed for both types of comparisons [73]. All permutation analyses were conducted using Rundom Project 1.1 [75], while the remaining linear morphometric analyses were performed using R software v4.0.4 [76].

To quantify both directional and fluctuating asymmetry using geometric morphometrics, the influence on the asymmetric component of shape variation was measured by considering the average side effect (population-level directional asymmetry), the individual side effect (individual fluctuating asymmetry), the measurement error, the sex effect (sexual dimorphism), and the size effect through a general and species-specific Procrustes ANOVA [77]. In this analysis, landmarks 1 to 18 were paired and compared with landmarks 37 to 20, with landmark 19 serving as the axis of asymmetry (the midpoint, without a corresponding landmark; Figure 3B). In the same way as linear morphometrics, we use two methods of performing the analysis of Procrustes ANOVA by geometric morphometrics: (1) using the software MorphoJ, which uses parametric significance tests and was specifically designed to assess the asymmetric component of the shape [78], and (2) using the Geomorph package in R through “bilat.symmetry” and “procD.lm” functions, which uses permutation tests to assess significance, and is widely used to assess the general shapes components [76,79].

3. Results

3.1. Directional Asymmetry

From the linear morphometrics approach, the frequency histogram of bilateral asymmetry values (trait) for the 72 individuals (clade level; Figure 4A) shows a unimodal normal distribution with a mean different from zero, resembling the expected curve for a trait exhibiting directional asymmetry (Figure 4B). The values tend to be negative, indicating that the left side of the individuals is generally larger than the right, a trend also visible in the bar plots for each species (Figure 4C), where the left side consistently appears larger. These results demonstrate the presence of left-sided directional asymmetry and rule out the presence of antisymmetry at both the species and clade levels. Statistically, both the Mann–Whitney U test and the permutation tests indicate a significant difference between sides (p < 0.05;

Table 1. Results of the Mann–Whitney U tests and the two-sample permutation tests for differences between sides (BA) at the clade and species levels. The p values indicate the significance of the tests.

Species	Wilcoxon	p	p (Permutation)
Clade	1697	<0.001	<0.0005
A. agassizii	263	0.29	0.31
A. cavernosus (F/M)	102	<0.01	<0.005
A. cordatus	311	<0.005	<0.005
A. cavernosus (P)	299	<0.01	<0.05

) at the clade level, supporting the findings from the histogram and bar plots. At the species level, A. cavernosus from Patagonia, A. cavernosus from the Falkland/Malvinas Islands, and A. cordatus all show significant differences between sides. In contrast, A. agassizii is the only species that does not show a significant difference between sides in either of the tests performed (Table 1).

The results of asymmetry differences (BA) between sexes, based on the Mann–Whitney U tests and permutation tests, suggest that there are no significant differences in the asymmetry trait (BA) between males and females at either the clade or species level (p > 0.05;

Table 2. Results of the Mann–Whitney U tests and the two-sample permutation tests for differences in asymmetry (BA) between sexes at the clade and species levels. The p values indicate the significance of the permutation tests. (P): Patagonia, (F/M): Falklands/Malvinas. The test was not performed for A. cavernosus (F/M) because the sample included only male individuals.

Species	Wilcoxon	p	p (Permutation)
Clade	739	0.243	0.27
A. agassizii	73	0.223	0.105
A. cavernosus (F/M)	-	-	-
A. cordatus	62	0.393	0.243
A. cavernosus (P)	62	0.37	0.34

). In contrast, both the Spearman correlation test and permutations tests conducted between individual asymmetry trait and size in the sample of 72 individuals show low and non-significant correlations at both the clade and species levels (

Table 3. Results of Spearman and permutation correlation tests at the clade and species levels. Rho and r indicate the correlation coefficients, while p values indicate the significance of the tests. (P): Patagonia, (F/M): Falkland/Malvinas.

Species	Rho	p	r	p (Permutation)
Clade	−0.072	0.51	−0.098	0.409
A. agassizii	0.046	0.840	−0.113	0.664
A. cavernosus (P)	−0.06	0.801	−0.117	0.621
A. cordatus	0.226	0.248	0.265	0.257
A. cavernosus (F/M)	−0.382	0.327	−0.311	0.375
Size range dataset	−0.072	0.51	−0.097	0.379

). Similarly, in the dataset of 81 A. cavernosus individuals from Patagonia, both the Spearman correlation and permutation tests again reveal a low and non-significant relation between size and asymmetry (Table 3; Figure 5).

From the geometric morphometrics perspective, the results of the Procrustes ANOVA conducted using MorphoJ reveal significant directional asymmetry (DA) both at the clade level and for each individual species (p < 0.0001;

Table 4. Results of Procrustes ANOVA using MorphoJ at the clade and species levels, assessing the effect on the shape asymmetric component of the individual, side (DA), ind*side (FA), species (only in the clade), sex, size, and population (only in A. cavernosus). (P): Patagonia, (F/M): Falkland/Malvinas.

	Effect	SS	MS	df	F	p
Clade	Ind	0.03015452	0.0000146027	2065	2.99	<0.0001
	Side	0.00460383	0.000131538	35	26.92	<0.0001
	Ind*Side	0.01214057	0.0000048855	2485	10.59	<0.0001
	Error	0.00232545	0.0000004614	5040
	Species	0.07307232	0.000695927	105	47.66	<0.0001
	Sex	0.00012825	0.0000036644	35	0.25	1
	Size	0.00500451	0.0000178732	280	1.22	0.0101
A. agassizii	Ind	0.01120998	0.0000200178	560	3.07	<0.0001
	Side	0.00069637	0.0000198964	35	3.05	<0.0001
	Ind*Side	0.00456341	0.0000065192	700	15.38	<0.0001
	Error	0.00062325	0.000000424	1470
	Sex	0.00043284	0.0000123668	35	0.62	0.9597
	Size	0.00212148	0.0000202045	105	1.01	0.4617
A. cavernosus (P)	Ind	0.00507296	0.000010353	490	4.96	<0.0001
	Side	0.00273576	0.0000781644	35	37.46	<0.0001
	Ind*Side	0.0013877	0.0000020868	665	5.04	<0.0001
	Error	0.00057915	0.0000004137	1400
	Sex	0.00029063	0.0000083038	35	0.8	0.7853
	Size	0.00091105	0.0000065075	140	0.63	0.9994
	Population	0.03793207	0.0010837736	35	103.34	<0.0001
A. cavernosus (F/M)	Ind	0.00184649	0.0000075367	245	2.46	<0.0001
	Side	0.00210736	0.0000602103	35	19.64	<0.0001
	Ind*Side	0.00107307	0.0000030659	350	5.93	<0.0001
	Error	0.00039803	0.0000005169	770
	Sex	na	na	na	na	na
	Size	0.00258541	0.000024623	105	3.27	<0.0001
A. cordatus	Ind	0.00550034	0.0000112252	490	2.14	<0.0001
	Side	0.00237164	0.0000677611	35	12.92	<0.0001
	Ind*Side	0.0034887	0.0000052462	665	10.07	<0.0001
	Error	0.00072941	0.000000521	1400
	Sex	0.00025461	0.0000072744	35	0.65	0.9422
	Size	0.00522161	0.0000372972	140	3.32	<0.0001

). The mean squares (MSs) indicate the magnitude of asymmetry, with species ordered from highest to lowest level of asymmetry as follows: (1) A. cavernosus (Patagonia), (2) A. cordatus, (3) A. cavernosus (Falkland/Malvinas), and (4) A. agassizii. In these analyses, there is no evidence of a significant influence of sex on the shape asymmetric component neither at clade nor species level (Table 4). In the case of size, this variable does not show a significant influence on asymmetry except for A. cavernosus (F/M) and A. cordatus (Table 4). Using the geomorph package in R, significant directional asymmetry is also detected at the clade level (

Table 5. Results of Procrustes ANOVA using Geomorph package at the clade and species levels, assessing the effect on the shape asymmetric component of the individual, side (DA), ind*side (FA), species (only in the clade), sex, size, and population (only in A. cavernosus). (P): Patagonia, (F/M): Falkland/Malvinas Islands.

	Effect	SS	MS	df	F	p	Distance
Clade	Ind	0.217697	0.0030662	71	6.298	0.7096
	Side	0.009216	0.0092157	1	18.929	<0.0001
	Ind*Side	0.034566	0.0004868	71	15.043	<0.0001
	Error	0.00466	0.0000324	144
	Sex	0.000489	0.00048913	1	1.1842	0.268
	Size	0.000183	0.00018345	1	0.4442	0.681
	Species	0.006633	0.00221095	3	5.3529	<0.001
A. agassizii	Ind	0.027574	0.00137872	20	2.5439	0.8866
	Side	0.001394	0.00139405	1	2.5722	0.0979	0.00814
	Ind*Side	0.01084	0.00054198	20	18.2423	<0.0001
	Error	0.001248	0.00002971	42
	Sex	0.0005157	0.00051571	1	0.9087	0.372
	Size	0.0001087	0.0001087	1	0.1915	0.893
A. cavernosus (P)	Ind	0.0125584	0.000661	19	3.1591	0.9737
	Side	0.0054737	0.0054737	1	26.1615	<0.0001	0.01654
	Ind*Side	0.0039753	0.0002092	19	7.2205	<0.0001
	Error	0.0011591	0.000029	40
	Sex	0.0001193	0.00011935	1	0.626	0.641
	Size	0.0006146	0.00061458	1	3.2233	0.032
	Population	0.0002286	0.00022865	1	0.832	0.434
A. cavernosus (F/M)	Ind	0.0088743	0.0008874	10	2.2198	0.9364
	Side	0.004217	0.004217	1	10.5485	0.014	0.01956
	Ind*Side	0.0039977	0.0003998	10	11.0416	<0.0001
	Error	0.0007965	0.0000362	22
	Sex	na	na	na	na	na
	Size	0.000093	0.00009299	1	0.2143	0.88
A. cordatus	Ind	0.021988	0.0011572	19	2.4022	0.9437
	Side	0.004746	0.0047459	1	9.8513	0.0018	0.0154
	Ind*Side	0.009153	0.0004817	19	13.199	<0.0001
	Error	0.00146	0.0000365	40
	Sex	0.0001608	0.0001608	1	0.3188	0.805
	Size	0.0004168	0.00041682	1	0.8263	0.41

; p < 0.001). However, at the species level, only A. agassizii does not show significant directional asymmetry, consistent with the result obtained through linear morphometrics. When species are ordered from highest to lowest level of asymmetry based on mean squares, the same pattern observed in MorphoJ is recovered: A. cavernosus from Patagonia, A. cordatus, A. cavernosus from the Falklands/Malvinas, and A. agassizii. Additionally, the shape distance calculated using geomorph reveals a similar order, with a slight variation between A. cordatus and A. cavernosus from the Falklands/Malvinas. The order from highest to lowest shape difference between sides (DA) is as follows: A. cavernosus from Patagonia, A. cavernosus from the Falkland/Malvinas Islands, A. cordatus, and A. agassizii (Table 5). Finally, and similar to MorphoJ, in Geomorph the sex variable shows no influence on asymmetry at any level; however, size shows significance only for A. cavernosus (P) (Table 5).

3.2. Fluctuating Asymmetry

Using geometric morphometrics, both methods applied (MorphoJ and Geomorph) detect significant fluctuating asymmetry (FA) for each species (p < 0.0001). The mean square (MS) values for fluctuating asymmetry (ind*side) exceed the MS values for measurement error (Table 4 and Table 5), suggesting that there is no digitization error in the landmark data and that the observed FA can be attributed to factors other than measurement error. On the other hand, the Procrustes ANOVA reveals different levels of FA (effect of the individual’s side on the asymmetric component of shape) among the species. The mean square values for FA in both programs (MS for ind*side; Table 4 and Table 5) rank the species from highest to lowest FA as follows: (1) A. agassizii, (2) A. cordatus, (3) A. cavernosus from the Falkland/Malvinas Islands, and (4) A. cavernosus from Patagonia.

4. Discussion

This study quantified and characterized bilateral asymmetry in three species of Abatus using two morphometric approaches that apply different shape measurement methods. The main results reveal directional asymmetry in each of the three species, with a clade- and species-specific trend toward larger left sides compared to the right, and interspecific variation in the intensity of this asymmetry. In addition, significant fluctuating asymmetry was found, varying in intensity in an opposite pattern to directional asymmetry. These findings demonstrate the utility of combining both morphometric approaches for studying morphological patterns, particularly in relation to species symmetry.

The results suggest the presence of directional asymmetry in all three species, which may represent a morphological pattern rather than individual stochastic variation. However, the origin and/or function of the trait remains uncertain. The directionality in the body sides in Abatus aligns with reproductive traits such as the larger size of brood pouch IV (anterior-left) and the position of the gonopores, where two of the three are located on the left side of the body [41,59,80]. Commonly, sexual dimorphism in body shape (phenotype) may be influenced by both genetic and environmental processes [81]. In crustaceans (especially isopods with brood pouches) and chitons, differences in the shape of body regions critical for reproduction have been documented between sexes, often associated with seasonal or ontogenetic changes related to sexual maturation, gonad growth, or offspring incubation [35,82,83]. Based on this, we expected the trait in Abatus to vary between sex and size. However, no variation in asymmetry between sexes was evidenced in either linear or geometric morphometrics. On the other hand, while geometric morphometrics shows an influence of size on the asymmetric component of shape (dependent on the methodology used) in A. cavernosus (F/M and P) and A. cordatus, consistent with findings for A. cordatus in which oral plate asymmetry is influenced by individual size [57], linear morphometrics shows no correlation between asymmetry and body size (age). These findings suggest that body asymmetry in Abatus could not be related to a female reproductive trait, consistent with the absence of sexual dimorphism in the overall body shape of these urchins [62]. Moreover, the low correlation between size and asymmetry during ontogeny suggests that the trait behaves allometrically [84], with no higher asymmetry in adults compared to juveniles.

Other non-reproductive explanations may also account for the observed asymmetry in Abatus. Like many bilaterians, irregular sea urchins lack a symmetrical arrangement of internal organs, particularly the digestive system, which occupies most of the coelomic cavity [85,86]. It has been proposed that the accommodation of the digestive system, a long tube that loops twice around the body, could lead to the patterns of directional asymmetry observed in both extant and extinct species [57,87]. One hypothesis suggests a genetic control over body side expansion during growth to facilitate digestive system accommodation. Another proposes that asymmetry arises ontogenetically due to pressure from stomach contents (which are always full of sediment), causing one side of the test to expand [87]. Both explanations address the origin of directional asymmetry from genetic regulation or developmental constraint perspectives. However, the lack of correlation between size and asymmetry observed through linear morphometrics suggests that asymmetry may not result from ontogenetic pressure by the digestive system, as there is no increase in asymmetry with size or age. Nevertheless, evaluating asymmetry variation during ontogeny via individual morphometric tracking or assessing the heritability of the trait could clarify these hypotheses [26,88].

Test architecture in echinoids presents another potential explanation for the asymmetric patterns in Abatus. In spatangoids, there is high shape plasticity both phylogenetically and ontogenetically due to diverse processes occurring in test plates during growth [89]. Twenty parallel columns of plates arise from the peristome (mouth) and extend dorsoventrally in a radial fashion: five pairs in ambulacral zones and five pairs in interambulacral zones. The first pair of ambulacral plates are not identical and differ in size; the size of the plate varies by ambulacrum, producing alternating or offset patterns across plate columns. This feature is known as Lovén’s Rule and occurs mainly in echinoids, also causing non-homologous growth between the left and right sides of the body [90,91,92], potentially contributing to the directional asymmetry observed in Abatus. On the other hand, test plates grow at unequal rates during ontogeny, a process known as heterochrony [93,94]. Plates may also shift, change size, or follow different growth trajectories (plate translocation) [89], which confer a high plasticity of the spatangoids test. Furthermore, the different trajectories and growth of plates cause changes in mobility by affecting the amount and composition of spines and ambulacral feet on the test and have allowed their great adaptability and success in a variety of environments [93]. The lack of relation observed between this asymmetry trait and developmental constraints associated with sex or size allows us to hypothesize that the patterns of asymmetry could be related to the displacement of these sea urchins. However, studies of Abatus behavior in connection with its asymmetry are needed to clarify these assumptions.

Directional asymmetry has been documented in both vertebrates and invertebrates [95,96,97], and geometric morphometrics studies have shown that a subtle but statistically significant directional asymmetry of shape is nearly ubiquitous across animal taxa [1]. Although its intensity may vary depending on biological features (as sex, size or age) and ecological or biogeographic contexts (geographic origin/distribution or environmental variations), the presence of directional asymmetry should be regarded as a widespread condition rather than an occasional trait [1,98,99,100,101]. In Abatus, intraspecific variation in directional asymmetry has been reported by Stige et al. [57] in different A. cordatus populations in the Kerguelen Islands. Similarly, the results of this work show differences in directional asymmetry intensity between two populations of A. cavernosus, one continental (Magellan Strait, Patagonia) and one insular (Falkland/Malvinas Islands), which differ in body shape [62]. Dispersal capacity and gene flow in A. cordatus have been studied, revealing genetic differentiation across localities, even showing small-scale genetic structuring between patches just ten meters apart [102]. Thus, a genetic origin of DA in these species (to some extent) would align with the low dispersal capacity and patch-residency of Abatus, limiting gene flow and trait homogenization across populations and resulting in variation in asymmetry intensity. Under this assumption, the presence of the trait in different species could suggest a shared origin [97]. Although directional asymmetry has not been reported in other Abatus species, its presence in other bilateral echinoids, both extant and extinct [49,55,56], could suggest a phylogenetic inertia from a spatangoid ancestor.

The results of this study show significant fluctuating asymmetry in each of the evaluated species using both methodologies. The presence of fluctuating asymmetry alone is not an indicator of environmental stress at the sampling sites but rather reflects the natural tendency of organisms to deviate from the genetically and environmentally established phenotype, due to stochastic developmental noise, environmental shifts, or ecological interaction dynamics [1,10,12,103,104]. In this study, the highest levels of fluctuating asymmetry were observed in A. agassizii and A. cordatus, both sampled near human settlements, which has been linked to contamination of the sector and problems for A. agassizii [105]. Similarly, Mespoulhé [106], studying six populations of A. cordatus, reported developmental abnormalities in individuals near Port-aux-Français, possibly linked to wastewater discharge and chemical use for hydrogen production. Likewise, Lane [58] found morphological abnormalities in A. nimrodi at two sites near Casey Station in Antarctica that differed in human disturbance levels. Therefore, the elevated levels of FA observed in A. agassizii and A. cordatus may be associated with high anthropogenic disturbance in their habitats [105], aligning with lower FA levels found in A. cavernosus populations from less disturbed areas. This supports the potential of FA as an indicator of anthropogenic disturbance and developmental stress [12].

A comparison of DA and FA levels across Abatus species reveals an inverse relation between the two: species with higher DA tend to exhibit lower FA, and vice versa. This pattern may relate to the origin and characteristics of each asymmetry type. As previously mentioned, fluctuating asymmetry is defined as small, random deviations between the left and right sides, arising from stochastic developmental instability (due to internal or external factors) [11,22]. As these are subtle, non-directional, and individual-level processes, they can affect any part of a trait. Thus, when a directional pattern is present in a trait (DA), the irregularities caused by FA may variably affect the left–right difference in each individual, altering the trait’s population-level distribution and influencing DA intensity.

The results of this study confirm the presence of significant patterns of directional asymmetry in the three analyzed Abatus species, with a consistent tendency for the left body side to be larger, and with differences in DA intensity across species. While several possible origins and explanations of DA in this genus are discussed, further analyses and experiments are needed to clarify the causes of the observed pattern.