Inferring Developmental Modularity from Morphological Integration: Analysis of Individual Variation and Asymmetry in Bumblebee Wings

Abstract

Organisms are built from distinct modules, which are internally coherent but flexible in their relationships among one another. We examined morphological variation within and between two candidate modules: the fore- and hindwings of bumblebees (Hymenoptera: Apidae: Bombus empatiens). We used the techniques of geometric morphometrics (Procrustes superimposition) to analyze the variation of landmark configurations in fore- and hindwings. Regression was used to correct for size-related shape variation (allometry). Principal component analysis revealed patterns of variation that were remarkably similar for individual variation and fluctuating asymmetry (FA). Because covariation of FA among parts must be due to direct transmission of the developmental perturbations causing FA, this agreement of patterns suggests that much of individual variation is also due to direct developmental interactions within each developing wing. Moreover, partial least squares analysis indicated that the patterns of shape covariation between fore- and hindwings were nearly the same as the patterns of within-wing variation. Shape covariation of FA was only found in bees that had been reared under elevated CO(2) concentration but not in bees from the control treatment, suggesting that the mechanisms of developmental interactions between fore- and hindwings are related to gas exchange. We conclude that the fore- and hindwings are developmental modules that maintain internal coherence through direct developmental interactions and are connected to each other only by relatively few links that use the system of interactions within modules.

Evolutionary Persistence of Phenotypic Integration: Influence of Developmental and Functional Relationships on Complex Trait Evolution

“…Yet, it is rarely examined whether the source of trait correlations (i.e., development or function) affects the historical persistence of an integration pattern. Developmental integration is produced when components of a complex structure share common developmental precursors, pathways, or resources and it can be environment independent (Riska 1986;Atchley and Hall 1991;Cheverud 1995;Raff 1996;Klingenberg and Nijhout 1998;Klingenberg et al 2001;Schwenk 2001;Hall 2003;Hallgrímsson et al 2003;Badyaev 2004;Badyaev and Young 2004). However, this conventional definition of developmental integration also includes correlations among traits that result from a similar response by individuals or species to comparable selection pressures and, thus, may reflect common evolutionary history, promoting the view that developmental constraints are equivalent to phylogenetic constraints (Lauder 1981; ''local constraints'' sensu Smith et al 1985;Gould 1989;McKitrick 1993;Watson et al 1995;West-Eberhard 2003).…”

Asimetría Fluctuante: Una herramienta morfo-funcional para Medir Estabilidad del Desarrollo

“…Estudiar el desarrollo de rasgos correlacionados, puede llegar a ser una de las formas más complejas de medir variables (eg. Klingenberg & McIntyre, 1998;Mardia et al, 2000;Klingenberg et al, 2001). En estos ejemplos particulares, los autores analizan la variación de la forma global del ala del insecto y como su asimetría puede ser relevante en el comportamiento del vuelo.…”

Lack of response to selection for lower fluctuating asymmetry of mutant eyespots in the butterfly Bicyclus anynana

“…There was no significant heterogeneity across lines, traits, generations and sexes in measurement error (using methods described by Van Dongen et al, 1999a, b), while measurement error itself was very low (using the two-factor ANOVAs of Palmer (1994): measurement error variance significantly lower than the side  individual interaction variance). The coordinates of the landmarks were also used to calculate the centroid size, as a measure of wing size (Klingenberg et al, 2001). This is the square root of the sum of squared distances from a set of landmarks to their centroid (references and details in Klingenberg and McIntyre, 1998).…”

“…In each generation, males, females and lines were analyzed separately. As all eyespots are formed by the same developmental mechanisms, the sequential Bonferroni procedure was applied to each set of related tests to avoid making type I errors (Leamy, 1999;Klingenberg et al, 2001). …”

“…FA may turn into DA during artificial selection experiments (Graham et al, 1993). Insect wings usually exhibit DA of size and shape, and have been shown to provide information on DS as a significant amount of phenotypic variation in DA is of environmental, rather than genetic origin (Brakefield and Breuker, 1996Breuker, , 2003Klingenberg and McIntyre, 1998;Windig and Nylin, 1999;Klingenberg et al, 2001). The studies of the insecticide-resistant strains of the Australian sheep blowfly Lucilia cuprina have shown that antisymmetry may also arise from increased FA (McKenzie and Clarke, 1988).…”