Asymmetries Among a Particular Species of Turtles


Please note! This essay has been submitted by a student.

Download PDF


This study determines for the first time the variations in the patterns of fluctuating (FA) and directional (DA) asymmetries in M. leprosa using geometric morphometric techniques. For this purpose, a total of 19 landmarks were putted on pictures of ventral aspects of 41 young dead specimens (20.0-76.9 mm). Symmetric and asymmetric components of shape were studied. For the former, the regression versus size was highly significant (P<0.01), and the size explained 35.51% of the shape changes. For the latter, it was showed that both FA and DA were present. FA was centered in medium plastron scutes (pectoral and abdominal). DA accounted for more variation than FA and was mainly centered in abdominal, femoral and anal scutes. Stability of both components was reached at about 70 mm of length. A significant FA would be explained by environmental poor conditions, such as pollutnats. Although biological explanations of detected DA are purely speculative, one possible biological explanation involved male-male lateralized, with Mauremys leprosa males attacking with the same front side and defending with the same limb, as well as a righting lateralized response. All these hypothesis needed to be tested I further studies.

Key words: geometric morphometrics, directional asymmetry, Ebre river, fluctuating asymmetry, Gompertz model, plastron

Essay due? We'll write it for you!

Any subject

Min. 3-hour delivery

Pay if satisfied

Get your price


During the developmental history or ontogeny of an organism, the modellation of its morphology is greatly influenced by the environment. Ontogenetic development of individuals can be altered resulting in asymmetries. Asymmetry is defined as a deviation of an organism (or a part of it) from perfect symmetry of bilateral characters and is composed by different categories (Cocilovo, Varela, & Quevedo, 2006). Fluctuating asymmetry (FA) occurs when there are minor differences in the degree of symmetry (Davis & Grosse, 2008). Directional asymmetry (DA) occurs when one side of a bilateral trait always develops more than the other side. FA represents a sensitive indicator of developmental stress and is a widely used parameter to evaluate instability and plasticity caused by stressful conditions during the development of organisms. Rather than being genetically determined, FA affects the ontogeny of individuals, being mostly influenced by environmental characteristics so increased FA is a result to disease, invasive species, off-road vehicles, pollutants, habitat fragmentation and nutritional stress (Davis & Grosse, 2008) (Buicǎ & Cogǎlniceanu, 2013) (McCall, 2014). As a result, FA has been promoted as a useful measure of the degree of population adaptation, making it an important tool for studying the well-being at both the individual and population levels. While DA can result from normal development, FA occurs when an organism deviates from normal development. DA has been detected in multiple groups of reptilian (McCall, 2014).

Some researches target amphibians and reptiles for asymmetry research but fewer study testudines. Turtle plastron has repeated structures at both sides of the median plane so characterizing normal patterns of asymmetry in plastron is important to recognize features that depart from expected ranges, although some asymmetries have been described even under normal conditions (Ayres Fernández & Cordero Rivera, 2004) (Davis & Grosse, 2008) (Rivera & Claude, 2008) (Băncilă, Plăiaşu, Tudor, Samoilă, & Cogălniceanu, 2012) (McCall, 2014). Given the need for this information, the present study determines the body shape variation of Mediterranean pond turtle (Mauremys leprosa leprosa Schweigger, 1812). In Europe, M. leprosa is common in S Spain and Portugal and along the Mediterranean coast of SW France. M. leprosa is under increasing stress due to habitat alteration, industrial and agricultural pollution, marsh drainage, aquifer water extraction, and fisheries bycatch, while minor threats are harvest for trade for consumption and pets (van Dijk et al., 2004) (Rodríguez-Rodríguez & Escrivà-Colomar, 2016). At the European level, it is considered vulnerable, while at the European Union level, it is a species of community interest that requires the designation of special areas of conservation (European Commission, 1992).

This study determines for the first time the variations in the patterns of FA and DA in a total of sixteen bilateral traits of M. leprosa. A at the same time the information gathered should provide a framework for future examinations of symmetry in turtle shells. The research has been done using geometric morphometric techniques, which allows a meticulousness into descriptions.

Material and Methods

Photographing Specimens

The study was conducted from animals dead by unknown causes in the recuperation center of “Canal Vell” in the Ebre National Park, located at the SW part of Catalonia, Spain.

A total of 141 young specimens (20.0-76.9 mm) were collected dead from the site mentioned above. None showed signs of injury or accessory scutes. Animals were photographed using a digital camera on their ventral aspect. Image capture was performed with a Nikon® D70 digital camera (image resolution of 2,240 x 1,488 pixels) equipped with a Nikon AF Nikkor® 28-200 mm telephoto lens and high resolution images were kept on file. The camera was placed on a tripod parallel to a copystand and photographed from above so the focal axis of the camera was parallel to the horizontal plane of reference and centered on the middle ventral aspect. Care was taken to ensure that the left and right sides were equally in view in each picture. A scale was included in the images to standardize each specimen size (mm unit). Sexes of the specimens were not gathered as they were too young. Detailed information of studied animals can be requested to first author.

Landmark Selection and Digitization of Sample Images

A total of 19 landmarks were digitized using tpsDig2 software version 1.40 (Rohlf, 2015) and then saved to a TPS file. The location of the landmarks is presented in Figure 2. Six plastron scute pairs of each specimen (gular, humeral, pectoral, abdominal, femoral and anal, each pair arranged bilaterally) were considered, as well as 7 in the median plane. The landmarks were chosen in order to have a good representation of the overall plastron shape and in a way that allow us to see important features of development and asymmetry and are based on previous studies, such those by (Davis & Grosse, 2008) (Băncilă et al., 2012) and (Barros, Resende, Silva, & Ferreira Junior, 2012). A digital measurement was the body length of each specimen, for which we measured the straight-line length of the carapace using the ruler (Regis & Meik, 2017). Each plastron was digitized twice by second and third authors independently to determine if our method produced repeatable estimates of asymmetry. Measurement error was estimated from a Procrustes ANOVA by considering individual as the main source of variation, nested by individual*side variation, variation in different image of a same specimen served as the first source of error, and residuals representing variation in digitized replicates as the second source of error (Cocilovo et al., 2006). The individual*side variation stand for FA and the side variation for DA (Cocilovo et al., 2006).

Symmetric Component of Shape

Superimposition of landmark configurations was performed by a Generalized Procrustes Analysis, which removes the spatial variation that does not correspond to form. A principal component analysis (PCA) with the covariance matrix of the symmetric component of the shape (i.e. differences among individuals) was performed to identify the major components of variation. We included information about size and shape through centroid size (CS) and Procrustes distance (PD) estimations. CS was calculated as the square root of the sum of squared distances of each landmark from the centroid of the landmark configuration was performed to detect how symmetric component of shape changes in relation to size. PD was calculated as the absolute sum of the distances between each landmark of one specimen and the configuration of the smallest specimen (Del Castillo, Segura, Flores, & Cappozzo, 2016). PD was plotted versus body length (BL), and fitted a nonlinear growth model (Gompertz y=a*exp(b*exp(cx)) (Hammer, Harper, & Ryan, 2001) to ascertain the 95% CI of the BL. This function has better statistical properties than those by von Bertalanffy y=a(1-be-cx) (Lutz & Musick, 1996) (Lutz & Musick, 1996). Another estimate was computed using regression on a linearized model using a linear regression of the symmetric component of the shape versus log CS.

Asymmetric Component of Shape

A Procrustes analysis of variance (ANOVA) was performed to study the asymmetric component of shape (multiple components might occur according to the symmetry of the object), which allows us to detect significance of different sources of variation, such as interindividual variation, FA, and DA. Besides, the following analyses: PCA and regression were performed on the asymmetric component of the shape, too. We used Procrustes ANOVA to quantify the amount of asymmetric variation; results are reported as sum of squares (SS) and means squares (MS) that are dimensionless.


The amount of measurement error was negligibly small compared to the source of variation dealt in the analysis (MS value for FA, compared to MS value for individuals, Table 1), so we proceed with all subsequent analyses (Table 1) using averaged replicas.

Symmetric Component of Shape

In the PCA, the PC1 and PC2 summarized 48.79% and 15.94% respectively (PC1+PC2=65.74%) of the explained variation. Along this axis, turtles that were located towards the positive region (smaller specimens) showed relatively wider plastron, specially on pectoral scutes. Turtles located towards the negative region of the axis (larger specimens) showed proportionally narrower pectoral and abdominal scutes and a shorter anal scute (Figure 2). The regression of symmetric component versus CS was highly significant (P<0.01), and the size explained 35.51% of the shape changes. Stability of this component was reached at about 70 mm of BL (Figure 3).

Asymmetric Component of Shape

The variation due to digitizing error represented only 5% of the total variance. Procrustes ANOVA showed that DA is present on this sample (P<0.01), as well as FA (Table 1). Particularly, DA accounted for more variation than FA. In the PCA of the asymmetric component of the shape, the PC1 and PC2 summarized 26.35% and 18.15% respectively of the variation (PC1+PC2=44.50%). Along this axis, turtles that are located towards the negative (larger specimens) region showed larger abdominal, femoral and anal scutes, and more deviated median line (Figure 4). Although the regression of asymmetric component against the log-transformed CS revealed that DA showed a significant increase during ontogeny (P=0.042) this ontogenetic shape change through the asymmetric component was markedly lower (1.42%) than those observed on the symmetric regression component of shape. Stability of this component was reached at about 70 mm of BL (Figure 5).


Landmark-based geometric morphometric analysis was used to describe the body shape variation that exists among M. leprosa samples.

There is evidence that inbreeding or suboptimal environments can lead to asymmetries in the number of turtle scutes (Davis & Grosse, 2008). The presence of fish ponds, application of feeds and communities dwelling along the Ebre Delta River, may possibly affect the physical and chemical condition of the river. This may be the reason why there is a significant FA. Size explained 35.51% of the shape changes for symmetric component, so we could expect shell to become increasingly asymmetrical as turtles age (Davis & Grosse, 2008). If plastron scutes grow larger by forming new tissue partly along the center suture line, the growth of one can impinge on the growth of the other. This may result in the fact that in older turtles, the suture line running down the center of the plastron tends to deviate from the straight line normally seen in younger turtles (Davis & Grosse, 2008). Our results showed that FA is centered in medium plastron scutes (pectoral and abdominal), our study’s results suggesting a different pattern that those detect by other authors. For instance, for yellow-bellied sliders (Trachemys scripta scripta) detected FA tended to be in the forward-most plastron scutes (gular and humeral) (Davis & Grosse, 2008). The differences between the FA results could be due to species or historic site differences. DA represented a percentage of total variance notably higher than FA.

Detected significant DA was mainly centered in abdominal, femoral and anal scutes. At this point, biological explanations of this pattern are purely speculative, however, one possible biological explanation would involve male-male competition for mate. If in many species male-male combat and male-male displays are initiated left side to left side (McCall, 2014), the pattern of large scutes on one side could have developed in the dominant side. If Mauremys leprosa males attack with the same front side and defend with the same limb then a DA could develop (i.e. “the Blacksmith’s arm”) (McCall, 2014). We suggest it would be also a righting response, when the animal rights itself over one side of the body after been overturned on the back, detected in other species such as Green turtle (Chelonia mydas) and Olive Ridley turtle (Lepidochelys olivacea) (Malashichev, 2016). These hypothesis would need be tested. And the fact we had no information on sex, due to the youngness of animals, does not allow to contrast male-females AD differences.

Size explained only 1.42% of the asymmetric shape changes for asymmetric component, which is a very low percentage, although the regression was significant (P=0.042). This confirms a slight despicable asymmetric change during growth, and the presence of asymmetry from early stages of development. A monotonic growth period is the most likely explanation, because a slight increase in symmetry was observed. Contrarily, a lack of relationship between FA and body size has been observed in studies on T. graeca (Buicǎ & Cogǎlniceanu, 2013) but Davis & Grosse (Davis & Grosse, 2008) found that FA increases with size and age, as our results show. On the other hand, we detected no differences on the length at which stability of symmetric and asymmetric components were reached (Gompertz model).

The results of this study represent an estimate of the fitness status of the studied Ebre population at present possibly be the result of the anthropogenic activities, although our results cannot determine what factor in the environment is the actual cause of the FA present within the current study’s sample sites. A similar study in the future, correlated with an estimate of the possible factors of human impact, could estimate is warranted.

In summary, our results indicate that Mauremys leprosa presents a subtle change in asymmetry during ontogeny. It would be relevant to assess a more diverse sample in the context of the changes in symmetry, in order to evaluate levels of flexibility of the plastron to evolve in response to environmental pressures. Further studies must focus on the analysis of DA and sexual dimorphism during the ontogeny in order to reveal if the patterns observed herein are stable and common in that species.

writers online
to help you with essay
banner clock
Clock is ticking and inspiration doesn't come?
We`ll do boring work for you. No plagiarism guarantee. Deadline from 3 hours.

We use cookies to offer you the best experience. By continuing, we’ll assume you agree with our Cookies policy.