Printer Friendly

Determination of Developmental Modules and Conservatism in the Fore- and Hind-wings of Two Species of Dragonflies, Orthetrum sabina and Neurothemis ramburii.



The wings of dragonflies are highly compartmentalized as shown by the major and minor veins separating the different compartments or modules. There is a long term hypothesis that compartments of the wings as bounded by the veins may correspond to units of "gene regulation". Are the different compartments 'units of gene regulation' and is there genetic conservatism on the wings of the dragonfly? This study was therefore, conducted to evaluate whether there is a number and pattern of developmental modules in dragonfly wings and determine whether there exists genetic conservatism based on intra and inter-modular variations in the wings. The study was conducted in two cosmopolitan species of Libellulid dragonflies. Different hypotheses were formulated and tested as to the possible spatial boundaries based on major wing venations.

A priori models applying the tools of geometric morphometrics were constructed and statistically tested for the goodness of fit test (GoF) statistic by comparing the observed and expected covariance matrices. Jackknife support values for each variational model were also computed using g as the GoF statistic. Results showed fair consistency in the observed number and patterns of hypothesized developmental modules implying that the wings of these species of dragonflies are highly conserved. It is concluded that there is genetic conservatism in the morphological spaces in the wings of the two species. (c) 2011 Friends Science Publishers

Key Words: Modules; Dragonflies; Fore-wing; Hind wing; Modularity


The wings of all dragonflies are divided into compartments by a system of netted veins. A number of studies on insect wings suggest that such compartments, including small parts of the wings, could be considered as autonomous units of morphological variation. Hence, each of them can be seen as separate developmental module (Cavicchi et al., 1981; Thompson and Woodruff, 1982; Cowley and Atchley, 1990; Cavicchi et al., 1991; Guerra et al., 1997; Pezzoli et al., 1997; Baylac and Penin, 1998; Birdsall et al., 2000; Zimmerman et al., 2000). Compartments in insect wings have also been hypothesized to correspond to distinct cell lineages and domains of gene expression (Garcia-Bellido et al., 1973; Lawrence and Morata, 1976; Lawrence, 1992). But how many modules are really found in the dragonfly wing and Are the modules morphologically integrated?

Morphological data have been used to characterize developmental, genetic, functional and evolutionary modules. Modules are sets of traits that are said to be internally integrated by interactions among traits but are relatively independent from other modules. Understanding the relationships between modules in the dragonfly wing can be informative about the underlying biological processes of compartmentalization in the wings. Since morphological structures are produced by developmental processes, understanding how development produces covariation between modules can have substantial implications for understanding genetic variation and the potential of the species for evolutionary change (Klingenberg, 2008).

In this study, we used morphological data in the wings of two species of dragonflies in determining autonomous units of morphological variation that could be considered as developmental modules. The approach consists of fitting alternative models of modularity to shape datasets with the purpose of finding the modules that best account for the co- variation structure in each dataset. In this approach, modules are defined as subspaces embedded within the overall phenotype space. Because the processes that produce those modules can overlap spatially, it is expected that modules show some degree of covariation, even if the processes responsible for them are effectively independent (Marquez,2008). The patterns of variational modularity and integration are assessed by testing alternative a priori models, each of which hypothesizes a distinct modular structure caused by specific functional or developmental mechanisms (Marquez, 2008).


Collection and processing of samples: Samples of dragonflies (Orthetrum sabina and Neurothemis ramburii), were randomly collected from different areas in Iligan City, Lanao del Norte, Tacurong, Sultan Kudarat and Zamboanga del Sur (Fig. 1). The wings were removed and placed in clear glass slides for scanning. Digital images were acquired from both sides of the fore-and hind wing using a HP 2400 scanner at 1200 dpi.

Analysis of the O. sabina and N. ramburii modules: Geometric morphometric tools that include landmark-based and procrustes methods were used to extract shape information from coordinate data in a set of steps that eliminate reflection and variation in scale, position and orientation (Rohlf, 2002, 2004, 2006, 2007 and 2008). The coordinates of 180 landmark points around the contour of the wings and the main veins were used as data.

A total of 10 a priori models were constructed for both fore-and hind wings, which generated 4 alternative models for the fore-wing and 1 alternative model for the hind wing. This is already inclusive of the null model, which assumes the absence of modularity and integration. Each model hypothesizes a distinct modular structure caused by a specific functional or developmental mechanism (Marquez, 2008). As shown in (Fig. 2 and 3), model 1 represented the null model; the second, third, fourth and fifth a priori models were set based on vein positions and wing compartments (Table I). Defined models were mixed by the software to produce variational or alternative models for analysis.

To test the acceptability of hypothesized modules (Figs. 2, 3 and 4) MINT ver. 1.0b software was used. MINT assumes that the data themselves have modular structure, and by partitioning the entire data space into orthogonal subspaces, covariance matrices were computed based on the modified data structure.

Estimating goodness of fit: The goodness of fit (GoF) tests, were employed to assess whether a pre less than defined model or hypothesis will be good enough to explain variation in a dataset. In this approach, the g values were used to determine whether the best-supported model is the one from which the data was derived when fitted to the original set of 5 models each for the fore-and hind wing and to the complete set of 9 possible model combinations for the fore-wing and 6 for the hind wing respectively (Figs. 2 and 3). The lower the g value simply indicates high degree of similarity between the observed data and the proposed model. Meanwhile, a low ( less than 0.05) P-value corresponds to large values of g , indicating a large difference between data and model and thus a poorly fitting model (Marquez, 2008). Determining model support: The best-fitted models were determined using jackknife support.

The jackknife support values for each model were computed by resampling 1000 replicates using g as the GoF statistic, dropping 10% of the specimens per jackknife replicate and computing 95% confidence intervals for the statistic. Finally, a measure of model support called "jackknife support" was computed by counting the proportion of jackknife samples in which a model ranks first (i.e., has the lowest value of g ) (Manly, 2006; Marquez, 2008).


Results showed that both species of dragonflies share the same best fit model as supported by the standardized gamma value (g value), P-value and Jackknife support values. The best fit model for the wings of the two dragonflies is Model 3 and 4 for the fore-and hind wings, respectively (Table II and III). These models yielded P-values greater than a=0.05; this accepts the hypothesis that these proposed models and the observed data are not significantly different. Model 3 and 4 also acquired the lowest (g value) and a jackknife support value of 100%, which qualifies it as a best-supported model among the five a priori models hypothesized for all samples tested.

Table I: A priori developmental and functional modules of modularity tested in this study. Modules correspond to regions of the fore- and hind wings of dragonflies (Family: Libellulidae) as hypothesized



H0:###{1}###{1}###No modules###"Null" model, predicting absence of modular structure

H1:###{2}###{2}###Single module###The wing is considered as a single homogenous developmental module.

H2:###{3}###{3}###Two modules###The wing is divided into anterior and posterior compartments, the anterior portion encompass the

###area from the Costa (C) vein towards the Media anterior (MA) vein, this is inclusive of the

###Subcosta (Sc) vein (2'~" longitudinal vein), Pterostigma and the Radius and Media (R+M) (the 3~

###and 4th longitudinal veins) with branches Ri -R4 reaching the wing margin; the 1R2 and IRS are

###intercalary veins behind R2 and R3 respectively. The posterior portion comprises of the

###compartments from the MA vein towards the posterior edge margin of the wing where, Cubitus

###(Cu) (5th longitudinal vein, Cubitus posterior (CuP), Anal veins (IA) are covered.

H3:###{4},{6},{7},{9}###{4},{6} Three modules###The area from Costa (C) vein towards the Radius and Media (R+M) (the 3rd and 4th longitudinal

###veins) inclusive of the Subcosta (Sc) vein (Yd longitudinal vein) and Pterostigma, black portion

###near the wing tip is considered as one module, then the area from R+M vein to Media anterior

###(MA) vein served as another module; from the MA to the Anal vein (IA) of the posterior edge of

###wing margin inclusive, of Cu, CuP served as one module.

###{5},{8}###{5}###Four modules###The area form Costa (C) vein towards Subcosta (Sc) (211d longitudinal vein) serve as one module.

###Then, from Subcosta (Sc) toward the Radius and Media (R+M) (the 3" and 4th longitudinal veins)

###inclusive of Pterostigma, black portion near the wing tip is considered as another module. The

###area from R+M vein to Media anterior (MA) vein inclusive of Intercalary veins (IR) served as

###one module; from the MA to the Anal vein (IA) of the posterior edge of wing margin inclusive, of

###Cu, CuP served as one module.

The best fit model for the fore- wings (Model 3) hypothesizes that anterior and posterior compartments represent distinct modules such that the genes controlling each module was further hypothesized to affect the developmental and genetic modularity of the wings. The anterior portion of the wings encompasses all the compartments bounded by the veins Costa (C) and Media anterior (MA) and includes the veins Subcosta (Sc) vein, Pterostigma, Radius and Media (R+M), branches R1-R4 touching the wing margin; the IR2 and IR3 are intercalary veins behind R2 and R3, respectively. The posterior portion comprises of all the compartments bounded by the MA vein and the posterior margin of the wing. This compartment includes all the veins Cubitus (Cu) (5th longitudinal vein, Cubitus posterior (CuP) and the Anal veins (IA).

The best fit model for the hind wing (Model 4) divides the hind wing into three distinct modules. The first module is bounded by the Costa (C) vein, the Radius and Media (R+M) and encompasses the the Subcosta (Sc) vein and Pterostigma. The second module is bounded by the R+M and the Media anterior (MA) veins. The third module covers the area bounded by the MA, Anal vein (IA) and the posterior margin of the hind wing (Fig. 5 and 6).

The hypothesized modules may be viewed as units of gene regulation and the main veins serve as possible boundaries. The fair consistency of the observed number and patterns of hypothesized developmental modules imply that the wings of these species are highly conserved indicating genetic conservatism in the morphological spaces in the wings of the two species.

The genotype-phenotype map has a modular structure implies that pleiotropic effects tend to be restricted to discrete subsets of phenotypic traits (Wagner and Altenberg, 1996), otherwise known as variational modules, which are again recognized by strong statistical associations among their component traits, in contrast to their weak associations with traits in other modules (Olson and Miller, 1958; Cheverud, 1982; Magwene, 2001; Klingenberg, 2005; Marquez, 2008). Modularity of the genotype-phenotype map has become an important focus for empirical and theoretical research, largely, because it is viewed as a condition of evolvability i.e., the ability to produce selectively useful variation (Wagner and Altenberg, 1996; Marquez, 2008). In addition to being building-blocks, modules are also the characters and homologs that unite phenotypes through evolution (Raff, 1996; Wagner, 1996; Eble, 2005; Scholes, 2008).

This aspect of variational modularity describes the condition of varying in modular fashion (Eble 2005; Scho es, 2008) and results from the operation of several unique evolutionary process that operate on phenotypic development (Raff, 1996; Scholes, 2008). The prevailing theory is that pleiotrophy evolves under natural selection to match the patterns of functional and developmental interdependencies among traits (Reidl, 1977; Cheverud, 1982; Wagner, 1988; Ehrich et al., 2003; Pavlicev et al., 2008; Marquez, 2008). Nonetheless, although there is genetic overlap across modules, on average each module is unique both in its set of expressed genes and in the way these genes interact among themselves and with their environment (Callebaut and Rasskin-Gutman, 2005).

Although modularity is expected to contribute only to statistical associations within modules, covariances between modules are rarely equal to zero, which means that more than one pattern of modularity may be supported by the data and intermodular associations are sometimes explained by factors spanning all modules (Wright, 1932; Magwene, 2001; Mitteroecker and Booktein, 2007 and 2008; Marquez, 2008). This is reflected in the results of the analyses wherein the other models also explain the modularity of the dragonfly wing, albeit to a lesser extent when compared to Model 3 for the fore-wing and Model 4 for the hind wing.

Table II: Jackknife support andy-values computed for each of the 9 models of variational modularity of Orthetrum saUna and Neurothemis ramburll fore-wing. Only the top best fit models are tabulated for the male and female fore-wing

Species and site###N###Best ranked %JS y-Value P value


Oflhzetrum .caNna

LANAODELNORTE###12###3###67.1 0.39415###1

Iligan City

SULTAN KUIDARAT###6###3###100 0.24576###1

Tacurong City

ZAMBOANOADEL SUR###50###3###100 0.18144###1

Neurothemis ranthurli

LANAODELNORTE###60###3###100 0.19241###1

Iligan City


Or/lie/rum sabina

LANAODELNORTE###12###3###100 0.49139###1

Iligan City

ZAMBOANGA DEL SUR###50###3###97.4 0.35164###1

Neurothemis ramburii

LANAODELNORTE###16###3###100 0.23300###1

Iligan City

Table III: Jackknife support and y- values computed for each of the 6 models of variational modularity of Orthetrum sabina and Neurothemis ramburii hind wing. Only the top best fit models are tabulated for the male and female hind wing

Species and site###N###Best###%JS###y- Value###P value


On/ic/rum saUna

LANAO DEL NORTE###12###4###50.3 0.47745###0.536

Iligan City

SULTAN KUDARAT###6###4###100###0.42906###0.975

Tacurong City

ZAMBOANGA DEL SUR###50###4###93.1 0.28671###1

Neurothemis rumburii

LANAO DELNORTE###60###4###100 0.32504###1

Iligan City


Or/hz c/rum sabina

LANAODELNORTE###12###4###76.4 0.45926###0.858

Iligan City

ZAMBOANGA DEL SUR###50###4###73.7 0.39377###1

Neurothemis ramburli

LANAODELNORTE###16###4###100 0.296070###1

Iligan City

Observed covariance structure is expected to reflect the cumulative effect of a number of embedded dimensions (Orr, 2000; Mezey and Houle, 2005; Hine and Blows, 2006; Marquez, 2008) each of which correspond in turn to a specific set of genetic, developmental, or functional interactions (Wagner, 1988; Marquez, 2008). These interactions can vary and thus evolve semi-independently from other similar processes such that covariance structures can be hypothesized to diverge even if they maintain a conserved intrinsic structure (Klingenberg, 2005; Marquez, 2008).

The patterns of modularity and integration observed for the wings of O. sabina and N. ramburii may be explained by the collaborative interactions among: (1) developmental, genetic and functional modularity, resulting from processes occurring within extant individuals and populations; and (2) evolutionary modularity, resulting from the history of divergence among evolutionary lineages in an entire clade. These types of modularity are mutually influencing each other through various processes within individuals or within populations (Klingenberg, 2008).

Results also confirmed a number of studies suggesting that the compartments, or even smaller parts of the wing, are autonomous units of morphological variation, and consequently, each of them is a separate developmental module (Cavicchi, Pezzoli and Giorgi, 1981; Thompson and Woodruff, 1982; Cowley and Atchley 1990; Cavicchi et al., 1991; Guerra et al., 1997; Pezzoli et al., 1997; Baylac and Penin, 1998; Birdsall et al., 2000; Zimmerman et al., 2000). Moreover, the results are in conformity with a number of studies that each wing vein has its own identity yet interconnected to the others and that signals originating from compartment boundaries initiate regulatory interactions that subdivide the wing into series of sectors with discrete boundaries such that different sectors are distinguished by the expression of a different combination to initiate vein formation (Chew, 2009; Sturtevant and Bier, 1995; Biehs et al., 1998; de Celis, 1998; Lunde et al., 1998; Milan and Cohen, 2000).

The specificity of this process gives each vein its identity (Gonzalez-Gaitan et al., 1994) such that wing compartments bounded by veins had been promising candidates for being separate developmental modules. Hence, a module occupies a specific morphological domain and corresponds to a single "morphogenetic field" or they can be viewed as units of gene regulation (von Dassow and Munro, 1999; Gilbert et al., 1996). Thus, studies regarding the subdivision of the wings into anterior and posterior compartments have attracted special attention, because these compartments may correspond to distinct cell lineages and domains of gene expression (Garcia-Bellido et al., 1973; Lawrence and Morata, 1976; Lawrence, 1992).


The results of this study suggest that the fore- and hind wings of dragonflies might comprise of 2 and 3 developmental modules, respectively. These hypothesized modules may be viewed as units of gene regulation and the main veins serve as possible boundaries. The fair consistency of the observed number and patterns of hypothesized developmental modules imply that the wings of these species are highly conserved indicating genetic conservatism in the morphological spaces in the wings of the two species. It is recommended that further studies on the modularity and integration of other species of dragonflies be tested before any robust conclusion be made regarding the number of modules comprising each wing. Acknowledgment: The researcher would like to extend her heartfelt gratitude to PCASTRD-DOST for the scholarship grant and Dalayap family for helping her in the collection of samples.


Baylac, M. and Penin, 1998. Wing static allometry in Drosophila simulans males (Diptera, Drosophilidae) and its relationships with developmental compartments. Acta Zool. Acad. Sci. Hung., 44: 97- 112

Birdsall, K., E. Zimmerman, K. Teeter and G. Gibson, 2000. Genetic variation for the positioning of wing veins in Drosophila melanogaster. Evol. Dev., 2: 16-24

Biehs, B., M.A. Sturtevant and E. Bier, 1998. Boundaries in the Drosophila wing imaginal disc organize vein-specific genetic programs. Development, 125: 4245-4257

Callebaut, W. and D. Rasskin-Gutman, 2005. Modularity: Understanding the Development and Evolution of Natural Complex Systems. USA: Massachussettes Institute of Technology

Cavicchi, S., C. Pezzoli and G. Giorgi, 1981. Correlation between characters as related to developmental pattern in Drosophila. Genetica, 56: 189-195

Cavicchi, S.G. Giorgi, V. Natali and D. Guerra, 1991. Temperature-related divergence in experimental populations of Drosophila melanogaster. III. Fourier and centroid analysis of wing shape and relationship between shape variation and fitness. J. Evol. Biol., 4: 141-159

Chew, P., 2009. Insect Wings. Available:

Cowley, D.E. and W.R. Atchley, 1990. Development and quantitative genetics of correlation structure among body parts of Drosophila melanogaster. American Nat., 135: 242-268

De Celis, J.F., 1998. Positioning and differentiation of veins in the Drosophila wing. Int. J. Dev. Biol., 42: 335-343

Garcia-Bellido, A., P. Ripoll and G. Morata, 1973. Developmental Compartmentalisation of the wing disk of Drosophila. Nature (London), 245: 251-25

Garcia-Bellido, A. and J.F. De Celis, 1992. Developmental Genetics of the venation patterns of Drosophila. Annu. Rev. Genet., 26: 277- 304

Gilbert, S.F. J.M. Opitz and R.A. Raff, 1996. Resynthesizing evolutionary and developmental biology. Dev. Biol., 173: 357-372 Von Dassow, G. and E. Munro, 1999. Modularity in animal development and evolution: elements of a conceptual framework for EvoDevo. J. Exp. Zool. (Mol. Dev. Evol.), 285: 307-325

Gonzalez-Gaitan, M., M.P. Capdevila and A. Garcia-Bellido, 1994. Cell proliferation patterns in the wing imaginal disc of Drosophila. Mech. Dev., 40: 183-200

Guerra, D., M.C. Pezzoli, G. Giorgi, F. Garoia and S. Cavicchi, 1997. Developmental constraints in the Drosophila wing. Heredity, 79: 564-571

Klingenberg, C.P., 2008. Morphological Integration and developmental modularity. Annu. Rev. Ecol. Evol. Systematics, 39: 115-32 Lawrence, P.A., 1992. The Making of a Fly: the Genetics of Animal Design. Blackwell scientific, Oxford, UK

Lawrence, P.A. and G. Morata, 1976. The compartment hypothesis. Sym. R. Ent. Soc., London, 8: 132-149

Lunde, K., B. Biehs, U. Nauber and E. Bier, 1998. The knirps and knirps related genes organize development of the second wing vein Drosophila. Development, 125: 4145-4154

Manly, B.F.J., 2006. Randomization, Bootstrap and Monte Carlo Methods in Biology, 3rd edition. Chapman and Hall/CRC Press, Boca Raton, Florida

Marquez, E.J., 2008. A statistical framework for testing modularity in multidimentional data. Evolution, 62: 2688-2708

Marquez, E.J., 2008. MINT: Modularity and Integration Analysis tool for Morphometric Data. Version 1.0 b (compiled 09/07/08). Mammals Division, University of Michigan Museum of Zoology

Milan M. and E. Cohen, 2000. Subdividing cell populations in the developing limbs of Drosophila: do wing veins and leg segments define units of growth control? Dev. Biol., 217: 1-9

Pezzoli, M.C., D. Guerra, G. Giorgi, F. Garoia and S. Cavicchi, 1997. Developmental constraints and wing shape variation in natural populations of Drosophila melanogaster. Heredity, 79: 572-577

Porto, A., F.B. De Oliveira, L.T. Shirai, V. De Conto and G. Marroig, 2009. The evolution of modularity in the mammalian skull I: Morphological integration patterns and magnitudes. Evol. Biol., 36: 118-135

Richtsmeiner, J.T., V.B. Deleon and S.R. Lele, 2002. The promise of geometric morphometrics. Yearbook Physiol. Anthropol., 45: 63-91 Rohlf, F.J., 2002. Tps Software Series. Stony Brook, Department of Ecology and Evolution, State University of New York, 2002

Rohlf, F.J., 2006. Tps Dig Version 2.10. Department of Ecology and Evolution, State University of New York at Stony Brook, New York. Rohlf, F.J., 2007. Tps Relw Version 1.45. Department of Ecology and Evolution, State University of New York at Stony Brook, New York. Rohlf, F.J., 2004. Tps Spline Version 1.20. Department of Ecology and Evolution, State University of New York at Stony Brook, New York. Scholes, E., 2008. Evolution of the courtship pheonotype in the bird of paradise genus Parotia (Aves: Paradisaeidae): homology, phylogeny and modularity. Biol. J. Linnaean Soc., 94: 491-504

Sturtevant, M.A. and E. Bier, 1995. Analysis of the genetic hierarchy guiding wing vein development in Drosophila. Development, 121: 785-801

Thompson, J.N. Jr. and R.C. Woodruff, 1982. Polygenic analysis of pattern formation: interdependence among veins in the same compartment of the Drosophila wing. Genetica, 60: 71-76

Wagner, G.P. and L. Altenberg, 1996. Perspective: complex adaptations and the evolution of evolvability. Evolution, 50: 967-976

Zelditch, M.L., D.L. Swiderski, H.D. Sheets and W.L. Fink, 2004. Geometric Morphometrics for Biologists: A primer. Elsevier Academic Press, New York

Zimmerman, E., A. Palsson and G. Gibson, 2000. Quantitative trait loci affecting components of wing shape in Drosophila melanogaster. Genetics, 155: 671-683

Department of Biological Sciences, College of Science and Mathematics, MSU-Iligan, Institute of Technology, Iligan City, Philippines
COPYRIGHT 2011 Asianet-Pakistan
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2011 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Tabugo, Sharon Rose M.; Torres, Mark Anthony J.; Demayo, Cesar G.
Publication:International Journal of Agriculture and Biology
Article Type:Report
Geographic Code:9PHIL
Date:Aug 31, 2011
Previous Article:Control of Rhyzopertha dominica (Coleoptera: Bostrichidae) by Pulverized Leaves of Punica granatum (Lythraceae) and Murraya koenigii (Rutaceae).
Next Article:Management Practices against Cotton Mealybug, Phenacoccus solenopsis (Hemiptera: Pseudococcidae).

Terms of use | Privacy policy | Copyright © 2022 Farlex, Inc. | Feedback | For webmasters |