Advances in Agricultural Science 07 (2019), 02: 2232
Geometric morphometrics study of wing shapes in the Calliptamus barbarus (Orthoptera: Acrididae)
Moad Rouibah^{1*}, Amel Hamouda ^{1} and Nadjet Badache ^{1}
ABSTRACT
Calliptamus barbarus (Orthoptera: Calliptaminae) is a most polymorphic species with one (1S) or three (3S) femoral spot. In the previous studies, comparisons of the two forms were made. In previous studies, different authors have attempted to compare the two forms of this species based on the morphology of the inner surface of the posterior femur (red or orange), classical morphometry, isolation reproductive, sound production, enzymatic and biochemical characters and recently a phyllogenic study based on the use of mitochondrial DNA. In order to distinguish these forms between them we conducted a landmarkbased geometric morphometric analysis on right and left elytra via the Generalized Procustes Analysis (GPA). This method allowed us to quantify the asymmetry of the elytra and identify shape changes between the two forms. It is performed by three mathematical operations: rotation, translation and scaling using the TPS software. Eleven landmarks were chosen at vein intersections. The obtained results show that the landmarks 4, 5 and 6 had low amounts of variability, but in general the displacement of landmark 11 on 1 and 3S form of C. barbarus is not very important. Thus the modifications of the wing shape between forms are slight. The effect of sex (male, female) and chromatic polymorphism on the wing asymmetry were not significant. This means that there is only one kind of asymmetry – a fluctuating asymmetry, and it is entirely due to development stress.
Keywords: Generalized Procustes Analysis (GPA), C. barbarus, Landmarks, Wing polymorphism
Introduction
Amid Calliptamus genus, C. barbarus is the most polymorphic species. It is characterized by a chromatic polymorphism of the hind femora (femore red with three spots or orange with only one spot).
In Algeria, C. barbarus was found for two areas: Littoral region for the form with three femoral spots (3S), and extremely semiarid and dry inhabitants for the form with only one femoral spot (1S).
Different authors have studied this species in order to compare the two forms on the basis of morphometric characters (Clemente et al. 1987; Benzara 2004; Larrosa et al. 2004), females ovarioles and biochemistry (Benzara 2004), acoustic emissions (Larrosa et al. 2008) and sexual relationships (Larrosa et al. 2007). Our last research of this species is a phylogenetic study based on molecular data (Rouibah et al, 2016). In order to continue our investigations of this species, we conducted a geometric morphometric analysis on wingvein landmarks.
According to Clarke (1993), there are two types of asymmetry: fundamental asymmetry due to a specialization of one of the sides and fluctuating or occasional asymmetry due to developmental stress. In this study we investigate whether a geometric morphometric of wings approach has the potential to inform us on the influence of geographic and chromatic polymorphism on evolutionary transformations within population of C. barbarus.
This is a method that allows to eliminate differences in size between samples. It is also the field of analysis of the biological form (Slice 2007), allowing to characterize quantitatively, to analyze and to compare the biological form (Bookstein 1991). In term of the type of imput data three general types of morphometry can be distinguished: traditional morphometry, contourbased geometric morphometry and a landmarkbased geometric morphometric analysis. In our study we have chosen the latter. The explanation of the principle of the Procustes method will be provided, and then this approach will be applied in the current work.
Material and methods
Principle of the Procustes method
An insect wing is most often a flattened structure formed by a network of veins supporting a membrane composed of two segregated tegumentary layers. The homology of the different veins of the wings has long been discussed. However, venation as a whole has long been regarded as a homologous structure across different groups of insects (Hammilton 1972, Comstock and Needham 1998). In this study, the landmarkbased geometric morphometric method is applied to the wing shape of C. barbarus. The purpose was to quantify the asymmetry of the elytra of each individual. This can be determined by a fundamental asymmetry that may emerged as a result of specialization of one of the sides. It is the case generally of males of Orthoptera Ensifera where the mirror of the right elytra rubs against the teeth carried by the left elytra (Benfekkih, 2006). It also possible that there is a fluctuating asymmetry due to genetic and / or environmental stress during development (Palmer and Strobeck 1986; Clarke 1993; Frampton & Hardersen 1999).
Procustes Analysis is a technique for comparing forms.The goal is to eliminate nonconformational variations. After this step, the differences between the landmark configurations will therefore be solely due to the shape variation. This is done by means of a Generalized Procustes Analysis (GPA), based on the least squares method (Gower 1975). It is performed by three mathematical functions: rotation, translation and scaling (Figure 1). This step makes it possible to obtain a ‘consensus’ representing the average shape of samples. The Procustes residuals represent deviations of each samples from the consensus of landmarks and then Procrustes residuals can be analyzed with the methods of multivariate statistics.
Application of the method
To carry out this study, slides were used as a support for the flattening of the wings. Glycerin was used to fix these wings. A lens equipped with a digital camera was used for taking photographs. The samples have been collected in August 2014 in 3 regions: samples of 3S form of C. barbarus was collected in Texenna (36° 69’ N; 5° 77’ E); samples of 1 S form were collected in Kasr El Boukhari (2° 67ˊ E; 35° 86ˊ N) and Djelfa (34°40′ N; 3°15′ E). This work has been partially carried out within the Department of Zoology of the University of Murcia in Spain and at the Laboratory of Zoology at the Faculty of Science of the University of Jijel. A total of 38 specimens are used, of which almost half (8 males and 8 females) were the 3 S form. The others (10 males and 12 females) were the 3 S. The elytra were flattened and mounted between two blades and fixed with glycerine, then photographed and finally scanned and saved as a JPEG image file.


The coordinates of landmarks (L) are digitized on each image by using the TPS dig2 version 2.17 software (Rohlf, 2013).
These coordinates are then stored in a text file in the format required for TPSRelW version 1.11 softrware (Rholf, 1997). Landmarks were chosen by taking into account the difference observed in the median field between the wings of the two forms (Figure 2, 3). The wing veins nomenclature used in this study was proposed by GrauvogelStamm et al. (2000); Bethoux & Nel (2001); Petit et al. (2006). Landmarks were as follow (Figure 4):
L1: insertion point of M with (A Cu + P Cu1)
L2: insertion point of of P1Cu with P Cu2
L3: the point of distal divergence of A Cu and P M
L4: the point of distal divergence of A Cu and P Cu1
L5: proximal starting point of A Cu and P M
L6: point of distal divergence of M
– L7: proximal insertion point of A R with P R
L8: point of proximal bifurcation of P R
L9: insertion point of A SC with leading edge
L10: insertion point of A R with distal edge
L11: insertion point of M A with posterior border
The coordinates of 11 landmarks of the 38 elytra pairs allow us, after superimposition, to compute of the angles formed by the vectors of the landmarks of the right elytra and the left one. This calculation was conduct for each individuals, to test whether some (L) have a preferred angle of deformation (Figure 5). The displacement of the (L) of the right elytra by contribution to the left one is represented by a vector whose angle with respect to the horizontal axe and the Euclidean distance can be calculated as the follow. If the coordinates of the points (Li) left and (Li) right are respectively (x1, y1) and (x2, y2), then the Euclidean distance is: : So the angle is (y2y1) / (x2x1).
The TPSRelW version 1.11 program (Rholf 1997) allows to perform all the measured objects to the same centroid size and to superimpose them by the GPA method, in order to calculate the consensus. The components of relative warps on the first 2 axes are considered to compare the different individuals. The contribution of the different L to the overall variability was evaluated by calculating for each L the sum of the Euclidean distances of each individual with respect to the consensus. Then it is evaluated by comparing the sum obtained for each L with the sum of the set for the 11 L.
The centroid sizes of each elytra were measured using GMTP version 2.1 software (Taravati & Darvish, 2010). For this purpose, the center of gravity of each elytra is calculated from the 11 (L) by the average of all the coordinates of the points of this form. Then the average Euclidean distance between each L and the barycenter of the 11 L is evaluated (Figure 6). Finally, ANOVAs were performed using Statistica software version 10.0, considering sex and chromatic polymorphism as factors. The aim is to know is the difference between Euclidean distances and centroid sizes of different forms and sexes statistically significant.
– P> 0.05: no significant difference; P <0.05: significant difference; P <0.01: very significant difference; P <0.001: highly significant difference
Results and discussion
Procustean superposition
The dispersion of the 11 L of the 38 pairs of elytra is illustrated in Figure7.
Figure 7 shows that L8 had the one of the largest scatter. While L 4, 5 and 6 have a low amounts of variability. From the point of view of the percentage of deformation, it must be noted that the L8 is the most unstable, with a contribution of almost 13% of the total variability.
Moreover, the angles between the vectors of the set of 76 elytra pairs were calculated for each L. The distribution histogram of the angles is shown for the L1 in Figure 9. The histogram shows that the majority of the values hovering around zero. This means that there is no privileged deformation within the angles of the 11 L of the right elytra by contribution to the left.
Influence of sex and polymorphism on the wing asymmetry of C. barbarus
The projections of the 38 pairs of elytra on the first 2 axes of relative flexion are placed in Figure 10. These projections make it possible to calculate, in a twodimensional space, the Euclidean distances between each (L) of the right elytra and the left one. This makes it possible to estimate the measurement of the deviation leading to asymmetry.
From these last projections, there is a heterogeneity in the dispersion of the points on the morphospace. For example, points 7, 8 on one side and 21, 22 on the other side correspond successively to the right and left wings of a male and a female at 3S form. On the other hand, points 19, 20 on the one hand and 53, 54 on the other hand, correspond successively to the right and left wings of a female and a male at 1S form.
The patterns of the relative flexions of the elytra show the different variations of instability of the landmarks (Figure 11). It should be noted also a remarkable displacement especially for L8: Figures A, B and D, L10: Figures A, B and C: L11: Figures B, C and D: L9: Figures B and D, L7: Figure C, L2: Figure D and L4:Figure B and to a lesser extent for the others Ls.
Table1. Percentage of contribution of each L’s in total variability.

Statistical analysis
To explain the variation of distances for each individual, a twofactor ANOVA (sex and polymorphism) without interaction was performed. The result indicates that there is no significant difference between the Euclidean distances of elytra of sexes (p = 0.7113) and forms of chromatic polymorphism (p = 0.0814) (Table 2).
The centroid sizes of the elytra of the 38 individuals are compared between sexes and forms of polymorphism. The result indicates a high significant centroid size differences (p = 0.004 <0.01) between sexes (Table 3), but there was no significant difference (p = 0.195> 0.05) between the polymorphic forms.

Table 2. ANOVA of Euclidean distances of elytra according to sex and forms of chromatic polymorphism.
SC: sum of squares; DDL: degree of freedom; MC: mean of squares; F: Fisher statistics; P: Probability to meet the computed value of the statistic
Table 3. Results of ANOVA: the effect of sex ad polymorphic forms on the centroid sizes of elytra

In a general way, the displacement of the 11 landmarks of the C. barbarous wings in 1S and 3S forms is not significant. This means that the alar shape changes between the semiarid population (Kasr El Boukhari and Djelfa) and that of the wetland (Texenna) are small. These differences are apparently due to the existing deformation in the alar morphology of the median sector of the elytra in the two forms. On the other hand, it should be noted that sex has no influence on the displacement of the 11 landmarks and that the influence of the polymorphism is small. It is also clear that the size of the elytra present in the two forms depends only on sex, the females were larger than the males, while the polymorphism has no influence on this size. Benfekih (2006) in his study of Locusta migratoria, found that the landmarks which are located in the median field, namely L 4, 5, 6 and 7 are the most stable. On the other hand, L 10 (insertion point of subcostal vein ASC with the anterior border) is the most unstable. In addition, Petit et al. (2006) reported in their study on the subject of wing geomorphometry of some Acridinae, Gomphocerinae and Oedipodinae that L7 (insertion point of the anterior margin with ASC), L1 (insertion point of median M with Cu cubital veins), (L)5 (insertion point of ACu and PM with CuP2) and the L8 (insertion point of AR with PR) are the subject to the strongest displacements. For its part, Petit (2007) in its study on Chortipus corsicus and C. pascuorum (Acrididae: Gomphocerinae) reported that for both species, L2 (the point of distal divergence of ACu and PCu1), L4 (the point of distal divergence of M), L7 (insertion point of ASC with the anterior border) and L8 (the proximal insertion point of AR with PR) were the most varied landmarks. The asymmetry observed between the right and left elytra of the 38 studied individuals can be determined as a fluctuating asymmetry. Consequently, there is no influence of chromatic and geographic polymorphism on the wing asymmetry of C. barbarus.

Conclusion
To prove that the population of C. barbarus of Jijel region belongs to the 3S form (3 spots), it is essential to carry out a comparative geometric morphometric study between the two forms. For this purpose, specimens are collected from wet area: Texenna and then compared with those collected from semiarid regions: Kasr El Boukhari and Djelfa. The aim of this study was to investigate possible wing asymmetry between the two forms of this Orthopteran species. This asymmetry may have a fundamental nature, or it may be defined as a fluctuating asymmetry, that is a result of environmental stress. The benefit of geometric morphometric approach is to obtaining a twodimensional (2D), graphical and statistical representation of the set of individuals studied separately. This method allowed to observe and to compare the variability of the wing conformation between the populations of the two forms of C. barbarus. The results obtained made it possible to note first that the landmarks located at the level of the median field (4,5 and 6) are the most stable and superimposed in C. barbarus as well as in other locusts as it has been reported by various authors. Moreover, it was indicated that the displacement of landmarks is poorly represented between the individuals of the two populations. This means that the only type of asymmetry present between the two forms and this is a fluctuating asymmetry due to development stress.
References
Benfekih, L. 2006. Recherche quantitative sur le criquet migrateur (Locusta migratoria) (Orthoptera : Oedipodinoe) dans le sahara Algérien perspectives de lutte biologique à l’aide de microorganismes pathogènes et des peptides synthétique. Thèse Doctorat. Université de Limoge, 140p.
Benzara, A. 2004. Polymorphisme géographique de l’espèce Calliptamus barbarus (Costa, 1836) (Orthoptera: Acrididae) en Algérie. Thèse Doctorat. Ins Nat. Agro., El Harrach, Alger, 154p.
Bethoux, O. & Nel, A. 2001. Venation pattern of Orthoptera. Journal of Orthoptera Research, 10: 195198.
Bookstein, F.L. 1991. Morphometric Tools for Landmark Data: Geometry and Biology. Cambridge University Press, New York, 435p.
Clarke, G.M. 1992.Fluctating asymmetry: a technique measuring developmental stress of genetic and environmental origin. Acta Zol. Fennica. 191:3135.
Clarke, G.M. 1993. Patterns of developmental stability of Chrysopa perla L. (Nevroptera: Chrysopidae) in response to environmental pollution. Environmental Entomology, 22: 13621366.
Clemente, M.E., Garcia, M.D & Presa, J.J. 1987. Morphometric and pigmentary variation in Calliptamus barbarus (Costa, 1836) in relationship with environnement, and its taxonomic value, in Evolutionary Biology of Orthopteriod insects. Ed. Boccetti, Ellis Homood Ltd, Chichester, 184189.
Comstock, J. H. & Needham, J. G. 1998. The wings of insects (continued). The American Naturalist, 32: 413424.
Frampton, C.M. & Hardersen, S. 1999. Effects of short term pollution on the level of fluctuating asymmetry – a case of study using damselflies. Entomologia Experimentalis et Applicata, 92: 17.
Gower, J.C., 1975. Generalised Procrustes analysis. Psychometrica, 40: 3350.
GrauvogelStamm, L., Nel, A. & MarchalPapier, F. 2000. Nouveaux Orthoptères (Ensifera, Insecta) du Trias des Vosges (France). Acta Geologica Hispanica, 35 (12): 518.
Hamilton, K. G. A. 1972 – The insect wing, Part 3. Venation of the orders. Journal of the Kansas Entomological Society, 45: 145162.
Larrosa, E., Ggarcia, M.D., Clemente, M.E. & Presa, J.J. 2004. El comportamiento en cautividad de Calliptamus barbarus (Orthoptera:Acrididae). Memorie della Societa entomologica italiana, 82 (2): 615630.
Larrosa, E, Garcia, DM, Clemente, EM & Presa, JJ. 2007. Estudio comparado del comportamiento en cautividad de dos bioformas de Calliptamus barbarus (Costa, 1836) (Orthoptera, Acrididae). Anales de Biología. 29:61–73.
Larrosa, E, Garcia, MD, Clemente, ME & Presa, JJ. 2008. Sound production in Calliptamus barbarus Costa 1836 (Orthoptera: acrididae: Catantopinae). Annales de la société entomologique de France (NS). 44:129–138.
Palmera, R. & Strobeck, C. 1986. Fluctuationg asymmetry: measurement, analysis, patterns. Annal Review of Ecological systematics, 17: 391421.
Petit, D. 2007. Comment interpréter la variabilité de la longueur des élytres chez deux Acrididae Gomphocerinae des montagnes de Corse ? Conference: Insectes d’altitude, insectes en altitude. 1 Actes des premières rencontres entomologiques du Massif central. Parc naturel régional LivradoisForez et Société d’Histoire naturelle Alcided’Orbigny, France, pp: 16.
Petit, D., Picaud, F. & EL Ghadraoui, L. 2006. Géométrie morphologique des ailes des Acrididae (Orthoptera, Caelifera) : sexe, stridulation, caractère. Annales de la société entomologique de France, 42 (1): 6373.
Rholf, F.J. 1997. TPSRelW, relative warps vers. 1.11. Logiciel et manuel.Ecology & evolution SUNY at SUNY Stony Brook. Available on http://life.bio.sunysb.edu/morph/morphnet/tpsrelww32.exe
Rholf, F.J. 2013. TPSSuper relative warps vers 2.00. Logiciel et manuel. Ecology & evolution SUNY at SUNY Stony Brook. Available on http://life.bio.sunysb.edu/morph/morphnet/tpsrelww32.exe
Rouibah, M., LópezLópez, A., Presa, J. & Doumandji, S. 2016. A molecular phylogenetic and phylogeographic study of two forms of Calliptamus barbarus. (Costa 1836) (Orthoptera: Acrididae, Calliptaminae) from two regions of Algeria. Annales de la société entomologique de France: 52(2)7787.
Slice, D.E. 2007. Geometric Morphometrics. Anthropol, 36: 261281.
Taravati, S. & Darvish, J. 2010. GMTP Version. 2.1. Manual. Geometric Morphometrics Tools Package. Available on http://tenebrionidae.net/files/GMTPmanual.pdf