Time Constraint and Genetic (Phenotypic)
Variation in Wing Shape in a Damselfly
Along a Latitudinal Gradient
Meagan Tunon
Degree project in biology, Master of science (2 years), 2020 Examensarbete i biologi 30 hp till masterexamen, 2020
Biology Education Centre and Animal Ecology, Uppsala University Supervisor: Frank Johansson
1 ABSTRACT
This degree project examined the effect of time constraint on wing shape and phenotypic variation in wing shape along a latitudinal gradient in the damselfly species Lestes sponsa. Fore and hind wings from individuals originating from three different latitudes: North (66°N), Central (59°N) and South (54°N) Europe were treated with their native temperature and photoperiod. In addition, the north and south populations were treated with south and north conditions respectively, resulting in five groups in total. Morphometric analyses of the wings revealed a positive correlation between body mass and wing centroid size, along with a difference in wing shape between the groups. Forewings and hindwings from the northern group were broader and rounder than wings from the central and southern groups. Additionally, the wings from the transplant groups resembled those of the native group of their treatment, indicating a phenotypic plasticity in wing shape. Lastly, statistical tests of phenotypic variation revealed that variation was highest in relative warp 2 in the forewings and hindwings, and this warp represents the curvature of the wing both upwards/downwards or towards the inside/outside of the wing. These results on phenotypic variation indicate that even in a new or changing environment, L. sponsa could be capable of adapting to varying temperatures and environmental conditions. This study builds our understanding of how this damselfly species, and potentially insects, will be affected by current and future potential climate change.
2 INTRODUCTION
The development of flight is one of many amazing evolutionary phenomena. This aspect of evolution has been subject to a wide range of natural selection, through both biotic and abiotic factors (Salcedo et al., 2018, Norberg, 1994). Flight has evolved in many organism groups, such as birds, bats, insects and even in fish (Speakman, 2001). Since flight is quite complex there is not one wing size or type that is ideal for all populations or species. This has resulted in a large variation in wing morphology, shape, size and other attributes. For
example, many beetles have large foldable wings (Muhammad et al., 2009), while butterflies have extremely large and broad wings with highly varied colors and patterns (Brakefield & French, 1999). Thus, characteristics of flight of organisms are modified and constrained not only by environmental factors but also by things such as physiology, life history, and behavior (Klingenberg, 2010).
One characteristic of flight that is quite interesting is how it is associated with wing shape. When looking at wings, morphology can be described as the collective shape, structure, form and color. The difference between shape and morphology is that shape is one aspect of morphology. Shape could thus be described more accurately as strictly the geometry of the wing itself. Since shape is part of the morphology, it is not surprising that also wing shape itself is affected by both environmental factors and genetic architecture (Stewart & Vodopich, 2018).
3
“driven” in certain directions by selection, but is subject to trade-offs in favor of improving one ability, sometimes reducing survival or reducing another ability (Outomuro et al., 2016). For example, wing shape is subject to both natural and sexual selection, in that it affects multiple aspects of an individual’s life such as predator avoidance or mating success.
Outomuro et al., (2016) found that natural selection (survival) favored individuals with long and slender forewings and short and broad hindwings, and that sexual selection favored the opposite. In cases like this, it is possible that these effects will be antagonistic (Outomuro et
al., 2016). However, it is not only selection from biotic factors that could have an impact on
wing shape. The ideal wing shape for an insect could also be impacted by abiotic factors, for instance, by the temperature, type of biome, and other natural geographical conditions of the region in which the organism lives. For example, Hernandez et al. (2010) found differences in wing shape in a species of moth which were related to altitude, being that the wing shape of moths from a higher altitude was narrower than those from a lower altitude.
One major environmental factor that affects life history and morphology of many organisms, including insects is time constraints (Laurila et al., 2001, Lytle et al., 2002). Time constraints are caused by the length of the season that is available for development and growth, and usually increase at higher latitudes. There is usually a shorter time available for growth and development at higher latitudes. Thus, seasonal constraint has a large impact on growth and development, and hence might also affect the shape of an insect wing. It has been shown that time constraints along latitudinal gradient can have an impact on many aspects of an insect’s growth and development. These changes have been identified in some insects as a decreased growth rate, reduced larval development time and reduced overall body size with increasing latitude, and an increased growth rate in lower latitudes (Tang et al., 2017). In regions where conditions are ideal for an organism (such as higher temperatures and greater levels of sunlight), it is possible that an organism may develop more slowly or without as much constraint and have a different wing shape, compared to an organism that has harsher conditions and is forced to develop quickly. However, the effect of seasonal time constraint based on latitude is not well explored when it comes to how this affects wing shape.
4
2015). This study suggests that not only can geographical conditions impact the development, body size, and later success of an insect, but also the shape of the wing. It also suggests that not only is it important to look at comparisons of a species in its entire range, and not just a north-south or west-east gradient.
Since time constraints are stronger at higher latitudes it is interesting to examine how
latitudinal gradient affect genetic variation in wing shape. Knowledge on genetic variation is important in that it determines the adaptability of an organism to changes in the environment. Since time constraints are stronger in the north we should expect less genetic variation at northern latitudes since selection for optimal development and growth should be stronger. A study supports the theory that genetic variation is highest at lower latitudes (Adams & Hadly, 2013). However, there are conflicting results on this. Latitudinal studies have found that genetic variation for traits can sometimes be higher at more northern latitudes (Van Els, et al., 2012). Less time constraints could also be associated with a more heterogeneous environment and such environments might cause a higher genetic variation in organisms. Several studies support the claim that genetic variation increases in more heterogeneous environments (Yeaman et al., 2010, Huang et al., 2014). Thus, this suggests that genetic variation will indeed change with latitude, as changes in latitude imply changes in environmental conditions and selective pressure.
Many organisms increase their range and thus encounter novel environments, and we need knowledge on how such changes in range margins affect important morphology that in turn affects fitness. For example, will the wing shape induced by the novel environment be in the direction that favors mating or survival? Since time constraints might affect wing shape it is interesting to examine how wing shape changes along a latitudinal gradient and whether a longer and slenderer or shorter and rounder shape is favored at northern and southern latitudes respectively. In addition, investigation into what will happen with phenotypic variation in wing shape when a species encounters a new environment is required. Will variation increase or decrease in the population? Such knowledge is important for predicting success of organisms that migrate north during climate change.
5
identify whether introducing an individual to a different latitudinal environment (transplant) will have an effect on wing shape. The third goal is to examine genetic variation in wing shape along the latitudinal gradient and whether a novel environment result in more or less genetic variation in wing shape.
Based on how seasonal time constraint affects the overall mass, development time, and size in L. sponsa, it is hypothesized that the latitudinal difference will have an impact on the shape of the wing. Finally, based on previous studies, it is hypothesized that genetic variation, here analyzed through phenotypic variation, will either increase or decrease along the latitudinal gradient.
METHODS
The aim of this study was to determine the effect of time constraint on wing shape of individuals of the species Lestes sponsa. L. sponsa is otherwise known as the common spreadwing or emerald damselfly. It is a damselfly that has a fairly wide distribution latitudinally, from northern to southern Europe (Sniegula et al., 2016). L. sponsa inhabits aquatic areas and lays its eggs in bodies of water. When the eggs hatch, they have a larval stage of a couple of months, where they then emerge into the terrestrial environment. Since this insect has a wide latitudinal distribution and has been shown to differ in various ways based on natural and sexual selection (Sniegula et al., 2016), it is thus a good study species to examine this effect of latitudinal time constraint on wing shape.
6
were collected in southern conditions. The central region eggs were only raised in their native conditions. The setup therefore produced five different treatment groups. These were the native groups: southern (native), central (native), northern (native), and the transplant groups: southern (northern treatment) and northern (southern treatment).
Experimental setup
Eggs from females were collected as in Sniegula et al. (2016), and thereafter brought to the Institute of Nature Conservation PAS in Krakow, Poland. Upon arrival, the climate chambers had conditions simulating this time of the year (late summer) for their respective regions (south, central and north). After 2-3 weeks of native conditions, the chambers were set to simulate winter conditions by lowering the temperature to 5 ℃ and switching off the lights. The eggs were kept in winter conditions for 4 weeks. After this wintering period, conditions simulating spring were started, using a thermo-photoperiod of the day when temperature went above 12 degrees Celsius for each region: 24 April, 26 April and 12 April for north, central and south respectively. During the spring simulation, all of the larvae hatched after about ten days. Natural photoperiod and temperature conditions for each region were then simulated until all larvae had emerged. Light level and temperature were changed every week to simulate real changing environmental conditions.
The larvae were reared individually in round plastic containers (diameter 7 cm, height 4 cm) and fed daily with laboratory reared brine shrimp Artemia salina. Six offspring from each female was raised. However, only 1-2 individuals per female were used for estimates of wing shape. Adult body mass was also recorded.
Analysis
For shape analysis of wings, only males were used. In November 2019, the right forewing and hindwing were removed from male specimens. In cases where the right wings were damaged, left wings were used instead. These were placed on a piece of paper with the specimen’s original latitude, treatment latitude, and ID number. Photos of each wing were taken using a
7
Panasonic DMC-FZ300 camera mounted to a stand, with the camera facing straight down towards the wings. Photos were taken at the same angle and distance for every specimen to ensure that each image was to the same scale. In some cases, more than one photo
was taken to ensure best image quality. Upon completion of photos, extra photos were trimmed away and the images were renamed to organize them in a logical way according to their origin, treatment, and ID.
With the files ready, the photos of the wings were processed with the use of the programs tpsUtil version 1.79 (Rohlf, 2019) and tpsDig version 2.31 (Rohlf, 2018). Table 1 shows the number of wings measured for each group. Using tpsUtil, scripts were generated for the fore and hind wings separately. Subsequently, the wing photos were loaded into tpsDig through these scripts. To examine wing shape, landmarks were used for each photo. Thirteen landmarks were used to capture the overall wing shape (Figure 1). These landmarks were chosen since they achieve an overall picture of the wing shape while also being comparable to previous studies that have looked at wing shape in this species (Outomuro, 2016). One landmark, landmark 10, was used as a sliding landmark and was allowed to slide across its tangent line. This is because this landmark’s location was not always in the exact same spot on every wing. The same 13 landmarks were clicked on every wing photo, and this clicking process created a TPS file that listed the image name and landmark data for each specimen.
From this TPS file, the program tpsRelw version 1.70 (Rohlf, 2019) was used to extract various data from the clicked landmarks. tpsRelw performs a Generalized Procrustes Analysis (GPA) (Rolfh, 2019) and calculates superimposed landmark coordinates by using generalized least square superimposition method which scales each specimen into unit centroid size, rotates it
Table 1: Table showing the number of each wing type measured for each group in the study. The North Southern treatment refers to wings from individuals collected in the Northern region but treated in Southern conditions, and the South Northern treatment refers to individuals collected in the Southern region but treated in Northern conditions.
Group Forewings Hindwings
North Native 67 67
Central Native 67 65
South Native 70 69
North Southern Treatment 59 60
8
and translates the coordinates into alignment (Bookstein, 1991). The data output from the program includes the weight matrix and consensus shape data for each different treatment. The weight matrix is the rotation of the Procrustes-residual shape coordinates; they are a set of shape coordinates for which the sum of squared differences is the squared Procrustes distance between any two specimens. It was used to compare wing shape between the groups. In addition, centroid sizes for each wing were extracted with tpsRelw. Using tpsUtil, these TPS and NTS files with data outputs were converted directly to CVS file types to be used in RStudio (RStudio Team, 2015) for further analyses.
To analyze shape differences a MANCOVA analysis was done in RStudio to test for the effect of treatment, centroid size, and the interaction of these two factors on the weight matrix (wing shape). In other words, this test was done to see whether the five groups of wings (South, Central, North, South: North and North: South) differed significantly in wing shape. I also regressed body size (mass) and wing centroid size to check for correlation of these two variables. In addition to these statistical analyses, visualization of wing shape was done with the program tpsSplin version 1.22 (Rohlf, 2016), or thin-plate spline. The consensus wing shapes for each treatment were projected onto the consensus of all the wings overall, in order to observe differences and allow for comparisons in wing shape for each group. This was done separately for the fore and hindwings. Lastly, using the same tpsSplin program, vector images of each wing shape showing the tendency of each landmark to move in a direction were produced (Appendices A and B). Because of multiple statistical comparisons, I corrected for type I error by doing Bonferroni adjustments.
Phenotypic variation
9
on these two relative warp scores, comparing them within two groups. These groups were: North, Central and South natives together, and North, South, and both transplant groups together. This test allowed me to see if the variance in these groups was equal, or if one or more groups were significantly different in their amount of variation. Additionally, I looked at the individual variance for both relative warps, for both wings, in each group. I also visualized shape variation in relative warp 1 and 2 with deformation grids (Appendix C).
RESULTS
Regression analysis showed that centroid wing size and body mass had a positive significant linear relationship in fore and hindwings, suggesting that centroid size is a good proxy of body mass (Fig 2 & 4). MANCOVA tests were run for both the fore and hind wings on the effect of treatment, centroid size, and their interaction. This test was done between various groups for comparison. The following groups were tested together: north, central and south natives; north and south natives; north native and north southern treatment; south native and south northern treatment; and north southern treatment and south northern treatment. Statistical testing and visual observations reveal a significant difference in wing shape among the various treatments (Fig. 3 & 5, Table 2 & 3). In all cases, treatment had a significant effect on the weight matrix, i.e. wing shape of both fore and hindwings. In most cases, centroid size also had a significant effect on the weight matrix. However, the interaction between these two factors was not always significant for each treatment, for neither the fore nor hind wings (Table 2 & 3). Differences in shape among the different treatments were similar in both the fore and hindwings (Fig 3 & 5). Bonferroni correction did not affect these significant results qualitatively (Table 2 & 3). Forewings
10
The MANCOVA analysis on the forewings revealed a significant effect of treatment on the weight matrix for all comparisons (Table 2). In addition, the effect of centroid size on weight matrix was also significant for all treatments. The interaction between these two factors however was only significant in three cases, these being the comparisons between: between south native and south northern treatment, and between south native and north southern treatment. It is interesting to note the difference in shape compared to the consensus
wings for each treatment (Figure 3). The northern native wing is quite broad and broad in the central area. The central native wing is slimmer overall and is narrower at the base. The southern native wing is quite similar to the shape of the north native wing but has a shorter and stubbier base, and somewhat thinner wing tip. The transplant treatment resulted in a different wing shape compared to that of the native wing shape (Table 2). The consensus shape of northern ones raised within southern conditions was more similar to the native south
Figure 3: Images of the thin-plate spline projections from tpsSplin of each group on the consensus fore wing shape, showing the average wing shape for each group. The transformations were enhanced by a factor of 10 to stress the differences in shape.
11
wing shape compared to the northern wing shape: it had a slightly upward curvature and narrower tip. Similarly, the shape of the wings of the south individuals raised in the northern conditions was closer to the shape of the northern wings than that of the southern wing: it was broader and wide at the tip. These shape change differences are visualized in the produced vector images (Appendix A). Here it can be seen that landmarks on the outer wing tips move inward in northern wings and outwards in southern wings. Landmarks from those individuals raised in transplanted conditions support these patterns with regard to north and south wing shapes.
Hindwings
Fig. 4: Graph showing the hind wing centroid size in relation to adult body mass (g). This relationship was significant, with a p-value of <0.0001.
12
Analysis of the hindwings provided some conclusive results as well. In some ways, the results were similar to those of the forewings but in other ways they were different. Just as in the forewings, the treatment had a significant effect on the weight matrix, that is wing shape, in every group (Table 3). Centroid size as well proved to be a significant factor, except in the comparison between the north native wings and the north wings with the southern
treatment. The interaction between treatment and centroid size was only significant in two of the comparative tests: between the north native and north with southern treatment groups, and north native and south northern treatment (Table 3). The shapes of the hindwings differences between groups differed somewhat from those seen in the front wings (Figure 5). The
northern native wings are fairly straight while being slightly broader towards their middle compared to those in the south. The central native wings are very narrow throughout. The southern native consensus shape is curved upwards with a downward pointed tip. The wings of individuals from the north but treated with southern conditions exhibited a longer base with a long and narrow tip. Lastly, wings of individuals from the south but treated with northern conditions were broader overall and with a rounded, wider tip and thus closer in shape to northern wings compared to the southern ones. Regarding the vector images of each wing group, results are similar to those of the forewing, with greatest variation in landmark location among 1, 13, and several along the bottom edge of the wing (Appendix B).
13 Phenotypic variation
Relative warp 1 and 2 explained 37.4% and 14.14% and 36.14% and 14.43% of the shape variation in the forewings and hindwings respectively. The variance, i.e. the phenotypic variation) for each warp within the 5 groups are given in Tables 4 and 5. None of the phenotypic variance values for relative warp 1 were significantly different between groups based on the Levene’s test for homogeneity of variance (Table 6). In contrast, there was a significant difference between south, central and north regions in relative warp 2 for both fore and hindwings (Table 6), but not for the comparison among the four groups when the central region was excluded (Table 6). The significant result for relative warp 2, suggests that there is higher phenotypic variance for central and north region wings fore and hindwings
respectively.
Group RelWarp1 RelWarp2
1-Central Native 0.0002245977 0.00007861494
2-North Native 0.0002014326 0.00005476873
3-South Native 0.0001432512 0.0000372239
4-South Northern Treatment 0.0001777541 0.0001250498
5-North Southern Treatment 0.0001931345 0.00004292923
Group RelWarp1 RelWarp2
1-Central Native 0.0001290022 0.00003372548
2-North Native 0.0001933834 0.00009231836
3-South Native 0.0001516424 0.00005493768
4-South Northern Treatment 0.0002126237 0.00008268801
5-North Southern Treatment 0.0001299755 0.00005150524
Comparison Relative Warp 1 Relative Warp 2
Forewing Hindwing Forewing Hindwing
North, Central, and South 0.2104 0.2276 0.02912* 0.002218*
Excluding Central 0.3657 0.4213 0.4045 0.2713
Table 4: Table of variance values for relative warps 1 and 2 in the forewing for each group.
Table 5: Table of variance values for relative warps 1 and 2 in the hindwing for each group.
14
The changes in shape represented by each relative warp for each wing are presented in Appendix C. In the forewing, relative warp 1 seems to account for the leaning of the wing towards the inside or the outside. Relative warp 2 seems to explain the curvature of the forewing upwards or downwards. In the hindwing, relative warp 1 seems not to affect wing shape very much. In the hind wing, relative warp 2 seems to impact the width of the base of the wing, and the overall width.
DISCUSSION
There were two main findings in this study. First, from the statistical and visual analyses, it is clear that there are some significant differences between the groups for both the fore and hindwings. Differences in wing shape along latitudes have been found in several insect groups e.g. in grasshoppers (Bai et al., 2016) and flies (Gallesi et al., 2016). The general pattern is that insects living at higher latitudes are both larger in size and have larger wings than those from more southern latitudes. Second, the relative warp analysis revealed some interesting information on phenotypic variance.
Native groups
There were several differences among the three native latitude groups (south, central and north natives). One of the more interesting results was that of the wing width. Additionally, the movement of landmarks 8, 9 and 10 were completely opposite directions between the north and southern native groups for both fore and hindwings (Appendix A and B). Wing width seems to be thinner in the south native groups, thinnest in the central native group, and widest in the north native group. This trend was seen in both in the forewings and the
15
efficient flying, and therefore dispersal (Hassal, 2015). This could be due to the fact that at lower latitudes, insects have a wider range of habitat that they can persist in due to more beneficial environments, while at higher latitude, the need for dispersal is less and it is more important to be able to be able to fly efficiently. An alternative non-adaptive explanation for the observed difference along the latitudinal gradient could be that since time constraints affect development and growth, the genes that affect these life history traits are also involved in wing shape formation and thus cause difference in wing shapes along the gradient studied.
Transplant groups
In looking at the transplant groups, there are differences between the wings of those
individuals originating in the north but receiving the southern treatment, and those originating in the south but receiving the northern treatment. Wings originating in the north but receiving the southern treatment are narrow and have quite a pointed tip, while those originating in the south but receiving the northern treatment are wider and have a rounder wing tip. Thus, in general the shape of the wings of the group from the north but receiving the southern treatment resembles the same shape as the southern native group in that they are narrower throughout. The same can be said for the wings originating in the south but receiving the northern treatment, which approach a similar shape to the northern native group, being wider and having a rounder tip. Based on these comparisons, one could say that the environmental effect (temperature and photoperiod treatment) has had a greater impact on the shape of these groups’ wings than any kind of genetic effect. Thus, phenotypic plasticity rather than genetic difference seems to form the observed wing shape when larvae were transplanted.
16 Forewings and hindwings
Due to time constraints, the forewings and the hindwings were not compared to each other specifically. However, since they were examined separately, we are still able to see general trends for each wing type between groups. The results showed that both the forewings and hindwings exhibit the same shape trends along the latitude. However, in other studies it has been shown that damselfly fore and hindwings differ in that forewings can exhibit less pointedness at the wing tip, and also be thinner at the base of the wing, while also being longer and slenderer than hindwings (Bots et al., 2012). Since the fore and hindwings are capable of moving independently from one another, it could be that each wing has a different role in the mechanism of flight. Though not tested in this study, based on visual similarities within the groups, it can be said that fore and hindwing shape are affected similarly by environmental conditions, but see Outomuro et al. (2016). In other damselflies such as
Calopteryx it has been shown that hindwings are usually wider at the base and the tip
(Outomuro et al., 2011). The reason for this is that hindwings are used for displaying sexual selected traits. However, since L. sponsa show no coloration on the wings I did not expect this pattern in L. sponsa.
Phenotypic variation
17
conditions. It is possible that this could be explained by the debated “canalization” or
“developmental buffering” effect. These theories suggest that there are unknown mechanisms involved in certain organisms’ physiology that prevent variation from becoming
detrimentally large (Breuker, Patterson & Klingenberg, 2006). This effect has been researched in a study (Llopis-Belenguer et al., 2015), which found evidence for
developmental buffering on the size and shape of dorsal and ventral anchors in Ligophorus
cephali, a species of flatworm.
Closing remarks and future studies
My results suggest that fore and hindwings in Lestes sponsa exhibit plasticity in the face of new environmental conditions. These findings are important given the increased pressures and threats presented by climate change. Since L. sponsa exhibits plasticity in wing shape, it is possible that even if temperatures are modified in their range, the species will be able to adapt to the shifting conditions and environmental stress. It is important to remember that species are affected not only by environmental pressures but by sexual selection as well, and this mix of effects can make it difficult to predict changes in species morphology or
suitability to their changing environment. Furthermore, given the possibility of
developmental buffering on wing shape, it could be that there will a limit to how much L.
sponsa wing shape will be able to adapt. Further studies could include checking the
differences between fore and hindwings specifically. Additionally, it would be interesting to test the effect of temperature and photoperiod within an even greater latitudinal gradient to see whether the effects on wing shape are more pronounced the greater the distance between origin and treatment latitude.
Acknowledgements
I would like to thank my project supervisor Frank Johansson for being so helpful,
18 REFERENCES
Adams, R.I., Hadly, E.A. (2013). Genetic diversity within vertebrate species is greater at lower latitudes. Evol Ecol 27, 133–143. https://doi.org/10.1007/s10682-012-9587-x
Bai, Y., Dong, J., Guan, D. et al. (2016). Geographic variation in wing size and shape of the
grasshopper Trilophidia annulata (Orthoptera: Oedipodidae): morphological trait variations follow an ecogeographical rule. Sci Rep 6, 32680. https://doi.org/10.1038/srep32680
Bookstein, F. L. (1991). Morphometric tools for landmark data: geometry and biology. Cambridge Univ. Press: New York.
Bots, J., Breuker, C. J., Kaunisto, K. M., Koskimäki, J., Gossum, H. V., & Suhonen, J. (2012). Wing shape and its influence on the outcome of territorial contests in the damselfly Calopteryx virgo. Journal of insect science (Online), 12, 96. https://doi.org/10.1673/031.012.9601 Brakefield, P.M. and French, V. (1999). Butterfly wings: the evolution of development of colour
patterns. Bioessays, 21: 391-401. doi:10.1002/(SICI)1521-1878(199905)21:5<391::AID-BIES6>3.0.CO;2-Q
Breuker, C. J., Patterson, J. S., & Klingenberg, C. P. (2006). A single basis for developmental buffering of Drosophila wing shape. PloS one, 1(1), e7.
https://doi.org/10.1371/journal.pone.0000007
Dochtermann, N.A. (2011). Testing Cheverud's Conjecture for Behavioral Correlations And Behavioral Syndromes. Evolution, 65: 1814-1820. doi:10.1111/j.1558-5646.2011.01264.x Frazier, R. M. (2007). Alpine Insects: Physiology and Evolution in Cold, Thin Air. University of
Washington. http://hdl.handle.net/1773/5244
Gallesi, M.M., Mobili, S., Cigognini, R. et al. (2016). Season matters: differential variation of wing shape between sexes of Calopteryx splendens (Odonata:
Calopterygidae). Zoomorphology 135, 313–322. https://doi.org/10.1007/s00435-016-0309-8 Hassall, C. (2015). Strong geographical variation in wing aspect ratio of a damselfly, Calopteryx
maculata (Odonata: Zygoptera) PeerJ 3:e1219 https://doi.org/10.7717/peerj.1219 Hernandez, N. & Barragán, A. & Dupas, S.e & Silvain, J-F & Dangles, O. (2010). Wing shape
variations in an invasive moth are related to sexual dimorphism and altitude. Bulletin of entomological research. 100. 529-41. 10.1017/S000748530999054X.
Hoffmann, A., & Shirriffs, J. (2002). Geographic Variation for Wing Shape in Drosophila serrata. Evolution, 56(5), 1068-1073. Retrieved from www.jstor.org/stable/3061645 Outomuro, D. & Johansson, F. (2015). Bird predation selects for wing shape and coloration in a
damselfly. Journal of Evolutionary Biology. https://doi.org/10.1111/jeb.12605
Outomuro, D. & Johansson, F. (2011). The effects of latitude, body size, and sexual selection on wing shape in a damselfly. Biological Journal of the Linnean Society, 2011, 102, 263–274. Hoffmann A. A. & Hercus J. M. (2000). Environmental Stress as an Evolutionary Force. BioScience,
Volume 50, Issue 3, 217–226, https://doi.org/10.1641/0006-3568(2000)050[0217:ESAAEF]2.3.CO;2
19
Laurila, A., Pakkasmaa, S. and Merilä, J. (2001). Influence of seasonal time constraints on growth and development of common frog tadpoles: a photoperiod experiment. Oikos, 95: 451-460. doi:10.1034/j.1600-0706.2001.950310.x
Llopis-Belenguer, C., Balbuena, J. A., Galván-Femenía, I., & Rodríguez-González, A. (2015). Phenotypic Buffering in a Monogenean: Canalization and Developmental Stability in Shape and Size of the Haptoral Anchors of Ligophorus cephali (Monogenea: Dactylogyridae). PloS
one, 10(11), e0142365. https://doi.org/10.1371/journal.pone.0142365
Lytle, D.A. (2002). Flash Floods And Aquatic Insect Life‐History Evolution: Evaluation Of Multiple Models. Ecology, 83: 370-385. doi:10.1890/0012-9658(2002)083[0370:FFAAIL]2.0.CO;2 Marchetti, K., Price, T., & Richman, A. (1995). Correlates of Wing Morphology with Foraging
Behaviour and Migration Distance in the Genus Phylloscopus. Journal of Avian
Biology, 26(3), 177-181. doi:10.2307/3677316
Marden H., J. (1989). Bodybuilding Dragonflies: Costs and Benefits of Maximizing Flight Muscle. Physiological Zoology 62, no. 2: 505-521.
Muhammad, A., Park, H.C., Hwang, D.Y. (2009). Mimicking unfolding motion of a beetle hind wing. Chinese Science Bulletin 54: 2416. https://doi.org/10.1007/s11434 -009-0242-z Norberg R., Å. (1994). Swallow tail streamer is a mechanical device for self-deflection of tail
leading edge, enhancing aerodynamic efficiency and flight maneuverability 257 Proc. R. Soc.
Lond. B. http://doi.org/10.1098/rspb.1994.0119
Outomuro et al. (2016). Antagonistic natural and sexual selection on wing shape in a scrambling damselfly. International Journal of Organic Evolution (2016) 70. doi:10.1111/evo.12951 Rohlf, F. J. (2018). tpsDig2, version 2.31. http://life.bio.sunysb.edu/morph/.
Rohlf, F. J. (2018). tpsSplin, version 1.22. http://life.bio.sunysb.edu/morph/. Rohlf, F. J. (2019). tpsRelw32, version 1.70. http://life.bio.sunysb.edu/morph/. Rohlf, F. J. (2019). tpsUtil32, version 1.79. http://life.bio.sunysb.edu/morph/.
RStudio Team (2015). RStudio: Integrated Development for R. RStudio, Inc., Boston, MA URL http://www.rstudio.com/.
Salcedo, M. & Hoffmann, J. & Donoughe, S. & Mahadevan, L. (2018). Size, shape and structure of insect wings: 10.1101/478768.
Sniegula, S., Golab, M.J., Drobniak, S.M. and Johansson, F. (2016). Seasonal time constraints reduce genetic variation in life‐history traits along a latitudinal gradient. J Anim Ecol, 85: 187-198. doi:10.1111/1365-2656.12442
Sodini SM., Kemper KE., Wray NR., Trzaskowski M. (2018). Comparison of Genotypic and Phenotypic Correlations: Cheverud's Conjecture in Humans. Genetics. 209(3) 941-948. doi:10.1534/genetics.117.300630.
Speakman, J.R. (2001). The evolution of flight and echolocation in bats: another leap in the dark. Mammal Review, 31: 111-130. doi:10.1046/j.1365-2907.2001.00082.x
Stewart, S. S. and Vodopich, S. D. (2018). Environmental effects on wing shape and wing size of
Argia sedula (Odonata: Coenagrionidae). International Journal of Odonatology: Vol. 21,
20
Tseng, M. and Soleimani, P.S. (2018). Body size explains interspecific variation in size–latitude relationships in geographically widespread beetle species. Ecological Entomology (2019),44, 151–156
Tang J, He H, Chen C, Fu S, Xue F. (2017). Latitudinal cogradient variation of development time and growth rate and a negative latitudinal body weight cline in a widely distributed cabbage beetle. PLoS One. 2017;12(7):e0181030. doi:10.1371/journal.pone.0181030
Van Els, P. & Cicero, C. & Klicka, J. (2012). High latitudes and high genetic diversity:
Phylogeography of a widespread boreal bird, the gray jay (Perisoreus canadensis). Molecular phylogenetics and evolution. 63. 456-65. 10.1016/j.ympev.2012.01.019.
Yadav, R. and Chang, N. (2014). Effects of temperature on the development and population growth of the melon thrips, Thrips palmi, on eggplant, Solanum melongena,. Journal of Insect
Science, Volume 14, Issue 1, 2014, 78, https://doi.org/10.1093/jis/14.1.78
23
Appendix C: The extremes of wing shape for each relative warp, left (left) and right (right).
Forewing, Relative Warp 1
Forewing, Relative Warp 2
Hindwing, Relative Warp 1
