CC BY
14
УДК 595.772
M. A. Chursina
Wing Shape in the Taxonomic Identification
of Genera and Species of the Subfamily Dolichopodinae (dolichopodidae, Diptera)
Voronezh State University, Voronezh, Russia
Abstract. Characters of the wing morphology have significant importance in the systematics and taxonomy of the Dolichopodidae family, but there are only a few studies concerning the variation in wing shape of dolichopodid flies. The detailed analysis of interspecific and generic wing shape variation can provide data for the taxonomic studies, while understanding of the selective forces shaping wing morphometric characters is important for studying the pattern of their evolutionary change. A geometric morphometric analysis was carried out on 72 species belonging to 5 genera of the subfamily Dolichopodinae in order to determine whether wing shape can be successfully used as a character for taxonomic discrimination of morphologically similar genera and species. Canonical variate analysis based on wing shape data showed significant differences among the studied genera and species. The analysis revealed wing shape variation, most of which was associated with displacement of posterior crossvein and apical sections of veins CuAp R4+5 and Mj+2, the two latter of which exhibited a trend toward convergence. Discriminant analysis allowed for the correct genera identification from 74.50% to 91.58% specimens. The overall success for the reassignment of specimens to their a priori species group was 84.04% on average. The detailed analysis of the variation in wing shape in the subfamily and out group taxa revealed evolutionary trends, the functional significance of which is discussed further. Database of wings shape will be created which can be used for taxonomic diagnostics of the family representatives and to conduct the studies of morphometric characters for various taxonomic levels.
Keywords: Diptera, Dolichopodidae, wing, shape, geometric morphometric, interspecific variation, taxonomy, phylogeny, Dolichopus, Hercostomus, Poecilobothrus.
DOI 10.25587/SVFU.2019.69.25523
М. А. Чурсина
Форма крыла в таксономической идентификации родов и видов подсемейства dolichopodinae (dolichopodidae, diptera)
Воронежский государственный университет, г. Воронеж, Россия
Аннотация. Признаки морфологии крыла широко используются в систематике и таксономии семейства Dolichopodidae, однако изменчивость формы крыла долихоподид изучена недостаточно.
CHURSINA Maria Alexandrovna - Candidate of Biological Sciences, assistant of the Department of Ecology and Systematics of Invertebrate Animals of the Medical and Biological Faculty of Voronezh State University.
E-mail: chursina.1988@list.ru
ЧУРСИНА Мария Александровна - к. б. н., ассистент кафедры экологии и систематики беспозвоночных животных медико-биологического факультета Воронежского государственного университета.
Подробное изучение межвидовой и межродовой изменчивости формы крыльев может помочь в таксономических исследованиях, а понимание направления отбора данных признаков важно для построения филогенетических схем и изучения эволюционных тенденций. Изменчивость формы крыла 72 видов из 5 родов подсемейства Dolichopodinae была проанализирована методами геометрической морфометрии с целью определить, возможно ли использовать форму для таксономической идентификации морфологически сходных видов и родов. Канонический анализ переменных формы крыла выявил значимые различия как между родами, так и между видами. Анализ показал, что большая часть изменчивости формы крыла была связана со смещением задней поперечной жилки и апикальных частей жилок CuA Я4+5 и M]+2, из которых последние две продемонстрировали тенденцию к сближению. Дискриминантный анализ показал, что форма крыла позволяет правильно определить род от 74,50% до 91,58% образцов. Успешное выделение образцов в их априорные виды было произведено в среднем в 84,04% случаев. Детальный анализ изменчивости формы крыла в подсемействе с учётом внешней группы позволил выявить тенденции эволюционных изменений, функциональное значение которых обсуждается. В дальнейшем будет создана база данных переменных формы крыльев, которая может быть использована для таксономической диагностики двукрылых семейства и изучения морфометри-ческих признаков на различных таксономических уровнях.
Ключевые слова: Díptera, Dolichopodidae, крыло, форма, геометрическая морфометрия, межвидовая изменчивость, таксономия, филогения, Dolichopus, Hercostomus, Poecilobothrus.
Introduction
The Dolichopodinae species have a wide geographical distribution; they are particularly abundant in humid forest, shores of water bodies and others wet habitats. Dolichopodinae are characterized by the large male terminalia, mid and hind femora with strong apical bristles and dorsally setose antennal scape. The largest genera of the subfamily are Dolichopus Latreille, 1796 (about 600 species), Hercostomus Loew, 1857 (about 500 species) and Gymnopternus Loew, 1857 (about 100 species) [1].
Although reliable information on the phylogeny of the Dolichopodinae subfamily is limited, recent studies confirmed monophyly of a clade, consisting of Dolichopus and Gymnopternus, and their separate systematic position were supported [2, 3, 4]. The genera Hercostomus, Poecilobothrus Mik, 1878 and Sybistroma Meigen, 1824 have been placed in a sister clade. However, before it was shown that Gymnopternus is an evolutionary independent entity, European and Russian dipterologists considered the genus Gymnopternus a subgenus of Hercostomus [5]. A strong dorsal seta on the first segment of hind tarsi can be used as a discriminator between Hercostomus and Dolichopus species; however, Poecilobothrus and Gymnopternus species are not clearly different from Hercostomus.
There is a considerable interspecific variation within subfamily in term of wing morphology, namely in the relative position of distal parts of R4+5 and M+2. The terminus of these veins may be subparallel beyond bending M1+2 (many species of Hercostomus) or convergent apically in species of Dolichopus h Gymnopternus (Pollet, 2003). Diagnostic character of Dolichopus species is vein M1+2 beyond crossvein dm-m with obtuse to angular S-shape bend and sometimes with stub vein, while species of Gymnopternus are characterized by straight vein M1+2 and R4+5 with slight posterior curve in distal section. However, several representatives of Gymnopternus have straight distal parts of vein M1+2 and R4+5, for example, Gymnopternus metallicus (Stannius, 1831).
Similar situation had arisen with the genus Poecilobothrus. This genus is a group of relatively large flies with distinct dark spot on notopleuron, triangular cercus and well developed epandrial lobe of hypopygium [6]. Negrobov [7] proposed a classification under which he
reduced Poecilobothrus to subgeneric rank within Hercostomus. Recent studies confirmed that Hercostomus is a polypheletic group [8], and presently Poecilobothrus are known as a separate genus, characterized by distinct dark spot above the notopleuron, triangular cercus and medium or large size.
In recent years three species previously belonging to the genus Hercostomus were transferred into Poecilobothrus: P. caucasicus (Stackelberg, 1933), P. varicoloris (Becker, 1917) and P. chrysozygos (Wiedemann, 1817) (Grichanov, 2018). This fact had considerably complicated the diagnostic characters determination of the genus Poecilobothrus, because P. chrysozygos has trapezoidal cercus, while P. caucasicus and P. varicoloris do not have notopleural dark spot. Several species of the genus Hercostomus are characterized by the presence of similar spot, for example, H. convergens and H. daubichensis, and some species (H. appolo, H. phoebus) are relatively large.
This confusion is not unexpected because the genus Hercostomus has already for a long time been a "basket" for species that do not fit into the other dolichopodine genera [2]. The main morphological characters distinguishing between Dolichopodinae species are the morphology of male termilalia and male secondary sexual characters. As a rule these characters are often used as diagnostic at least in some cases [9], while females of closely-related species are often inseparable morphologically. There are also difficulties in using the key taxonomic hard-to-detect diagnostic characters, or characters, the intraspecific variation of which has not been quantified. These difficulties can complicate taxonomical identification and classification.
Because dolichopodid flies play an important part in forest and agro-ecosystems as natural enemies of pests and have considerable potential as bioindicators [10], the problem requires the development of new alternative approaches.
The results of recent studies indicated that morphometric exploring wing shape variation represents an effective approach for finding differences between taxa. Wing morphometric analysis has been conducted in taxonomical studies of flies from the following families: Tabanidae [11], Tephritidae [12], Psychodidae [13] and Caliphoridae [14]. The data resulting in geometric method in conjunction with traditional morphological characters, molecular and ecological data has the capability of being utilized in describing evolutionarily transformation within a character system [15].
In the present study we used geometric morphometric wing shape analysis to examine the differences between Dolichopodinae genera and species. This information is expected to allow for the determination of the diagnostic characters of the genera and species. The resulting data of shape variation has the potential to contribute to a better understanding of evolutionary transformation trends in the subfamily.
Materials and method
In total, 7752 specimens (3192 females and 4560 males) of 72 species of the subfamily Dolichopodinae were examined, representing 5 the most common dolichopodid genera in Palaearctic region.
Material examined (number of wings): Dolichopus acuticornis Wiedemann, 1817 (11 5 <), D. argyrotarsis Wahlberg, 1850 (11 <), D. austriacus Parent, 1927 (5 13 <), D. campestris Meigen, 1824 (12 6 <), D. claviger Stannius, 1831 (6 8 <), D. jaxarticus Stackelberg, 1927 (12 15 <), D. latilimbatus Macquart, 1827 (60 14 <), D. lepidus Staeger, 1842 (7 16 <), D. linearis Meigen, 1824 (14 18 <), D. linearis Meigen, 1824 (28 36 <), D. longicornis Stannius, 1831 (38 50 <), D. longitarsis Stannius, 1831 (50 46 <), D. meigeni Loew, 1857 (4 6 <), D. migrans Zetterstedt, 1843 (6 10 <), D. nataliae Stackelberg, 1930 (7 14 <), D. nigricornis Meigen, 1824 (4 14 <), D. pennatus Meigen, 1824 (9 28 <), D. plumipes Fallen, 1823 (26 98 <), D. plumitarsis Fallen, 1823 (28 62 <), D. ptenopedilus Meuffels, 1982 (14 62 <), D. remipes Wahlberg, 1839 (18 13 <), D. rezvorum Stackelberg, 1930 (11 ?, 16 <), D. ringdahli Stackelberg, 1930 (69 81 <), D. rupestris Haliday, 1833 (25 41
S), D. sabinus Haliday, 1838 (13 8 S), D. simius Parent, 1927 (19 36 S), D. simplex Meigen, 1824 (16 32 S), D. trivialis Haliday, 1832 (113 148 S), D. ungulatus (Linnaeus, 1758) (59 97 S), D. zernyi Parent, 1927 (21 21 S), Gymnopternus aerosus (Fallen, 1823) (114 142 S), G. angustifrons (Staeger, 1842) (44 16 S), G. assimilis (Staeger, 1842) (8 6 S),
G. brevicornis (Staeger, 1842) (8 28 S), G.celer (Meigen, 1824) (62 118 S), G. congruens (Becker, 1922) (6 10 S), G. metallicus (Stannius, 1831) (81 56 S), G. pseudoceler (Stackelberg, 1933) (5 7 S), G. ussurianus (Stackelberg, 1933) (20 19 S), Hercostomus albibarbus Negrobov, 1976 (10 8 S), H. apollo (Loew, 1869) (6 8 S), Hercostomus chetifer (Walker, 1849) (14 11 S), H. convergens (Loew, 1857) (24 22 S), H. daubichensis Stackelberg, 1933 (6 6 S), H. eugenii Stackelberg, 1949 (4 6 S), H. excisilamellatus Parent, 1944 (6 9 S), H. fulvicaudis (Haliday, 1851) (10 12 S), H. fugax (Loew, 1857) (11 15 S),
H. fulvicaudis (Haliday, 1851) (10 12 S), H. germanus (Wiedemann, 1817) (24 38 S), H. kedrovicus Negrobov, Logvinovskij, 1977 (10 13 S), H. longiventris (Loew, 1857) (8 10 S), H. nigriplantis (Stannius, 1831) (53 70 S), H. phoebus Parent, 1927 (8 11 S), H. pterostichoides Stackelberg, 1934 (9 8 S), H. rivulorum Stackelberg, 1933 (6 7 S), H. rohdendorfi Stackelberg, 1933 (6 8 S), H.s rusticus (Meigen, 1824) (13 12 S), H. udovenkovae Negrobov, Logvinovskij, 1977 (6 7 S), H. vivax (Loew, 1857) (12 8 S), Poecilobothrus caucasicus (Stackelberg, 1933) (9 27 S), P. chrysozygos (Wiedemann, 1817) (40 91 S), P. clarus (Loew, 1871) (8 5 S), P. comitialis (Kowarz, 1867) (6 11 S), P. nobilitatus (Linnaeus, 1767) (24 50 S), P. principalis (Loew, 1861) (11 17 S), P. regalis (Meigen, 1824) (80 221 S), P. varicoloris (Becker, 1917) (24 25 S), Sybistroma binodicornis Stackelberg, 1941 (11 15 S), S. crinipes Staeger, 1842 (2 12 S), S. obscurella (Fallen, 1823) (22 22 S), Sympycnuspulicarius (Fallen, 1823) (25 12 S).
Sympycnus pulicarius (Fallen, 1823) (subfamily Sympycninae) served as an outgroup taxon. Dolichopodinae specimens were taken from the collection of Voronezh State University (Voronezh). For the widespread species we selected specimens from as many localities as possible in order to cover the range of intraspecific variation.
We used species of the subfamily Sympycninae as an outgroup taxon for the following reasons. The morphological analysis revealed a close relationship between Dolichopodinae and Sympycninae [16]. The most complete study of phylogenetic relationships of Dolichopodidae was provided by Bernasconi and Lim with co-authors [8, 17]. As demonstrated by these analyses, sympycnine species made a separated and well supported clade.
Wings of each fly were removed from body and mounted on a glass slide and covered with a cover glass. The slides were photographed by means of a Levenhuk C NG microscopic camera. Over each photo, a configuration of 8 type I landmarks (fig. 1) was digitized using the tpsDig software.
Fig. 1. Wing of Dolichopus cilifemoratus male, showing landmarks used in the study
Then geometric morphomertic analysis was performed. Firstly, a generalized Procrustes analysis was conducted in several steps: landmarks configurations were scaled to a unit of centroid size for eliminating the impact of variation in wing size, superimposed so that the centroid of each had coordinates (0, 0) and rotated so that the distance between landmarks of all specimens become minimal. A new set of variables (Procrustes residuals) contained the shape information and were used as shape data.
The centroid size of each wing was calculated to characterize an overall measure of a wing. All morphomertic and statistical analyses of these sets of variables were performed using the MorpholJ software and Statistica 10 for Windows.
In order to prove evidence of the significant differences in wing centroid size and shape among the genera and species, a one-way analysis of variance (ANOVA) with a Tukey post-hoc test, and a multivariate analysis of variance (MANOVA) was undertaken. Principal component analysis (PCA) was performed as an ordination method to describe the patterns of wing shape variation and to calculate the positions of each taxa within the morphospace. The shape changes associated with principle axes were visualized through thin-plate spline technique [18]. Comparative analyses of Dolichopodinae flies with outgroup taxa allowed for the determination of plesiomorphic and apomorphic character states of wing shape.
We used canonical variate analysis (CVA) combined with discriminant analysis (DA) to evaluate the diagnostic characters of the genera and to examine a probability of each specimen belonging to its a priori group (genera). The percentages of correct classification were used to evaluate the discriminatoring power of wing shape.
Then we quantified the intraspecific variation in wing shape to determine whether wing shape can be an effective tool to separate taxa on species level. CVA and DA allowed the estimation of the differences in wing shape and classification success similar to the previous case.
The matrix of the morphometric data was analyzed heuristically with stepwise addition option using Mesquite software. The tree was reconstructed by squared-change parsimony. This method has been used extensively for displaying evolutionary change of morphometric traits.
Results
Highly significant differences in wing centroid size were observed both among genera (F = 16.3, P < 0.0001) and among species (F = 27.0, P < 0.0001) (results of ANOVA). The Tukey post hoc test revealed no wing size differences between Poecilobothrus and Dolichopus, Hercostomus and Gymnopternus.
The MANOVA results indicated that there are significant wing shape differences among genera (Wilks' Lambda = 0.0118, F = 1375, P < 0.0001) and among species (Wilks' Lambda = 0.0003, F = 117, P < 0.0001).
Intergeneric variation and taxonomic discrimination
Detailed comparison of landmark configurations after Procrustes superimposition made it possible to determine wing shape features of studied genera. Landmarks 1, 5, 6 and 7 had the largest displacements. The posterior crossvein (landmarks 6 and 7) had most proximal position in Sympycnus species (fig. 2), followed by Sybistroma. Dolichopus, Hercostomus, Gymnopternus and Poecilobothrus species had more distal crossvein. Apart from that, Sybistroma and Sympycnus species exhibited the most distal insertion point of R2 with costal vein.
Landmark and 5 (the insertion point of M3+4 with wing edge) also exhibited high variation among studied genera. The most distal position of this point was observed in Sympycnus species, Poecilobothrus possessed the most proximal position of this point. The other landmarks did not vary widely among dolichopodine genera.
PCA results in two principle components accounting for 70,02% of the overall wing shape variation. The first principle component (PC1) explained approximately 47,44% of total variability. As shown by thin-plate splines, PC1 reflects displacement of landmarks 5, 6 and 7 in such a way that relative area of m} section decreases along the axis (fig. 3). In other words,
Fig. 2. Landmark configurations of the five Dolichopodinae genera and one outgroup genus after Procrustes superimposition
Fig. 3. Scatter plot from the first two principle components for Dolichopodinae genera and one outgroup genus with the associated shape changes
the first axis (PC1) described variation from a wing with relatively large m] section (Sympycnus species) to a wing with decreasing m1 section (such species as Poecilobothrus varicoloris, P. caucasicus, Dolichopus claviger).
The second principle component (PC2) accounted for about 22,58% of the variance and showed displacement of landmarks 5 toward the distal part of the wing, in other words, PC2 described a variation from a wider wing with shorter m1 (for example, Hercostomus rivulorum, Poecilobothrus comitialis) to an elongated wing with extended m1 (Sympycnus).
Two first canonical variates (CV1) accounted for about 84% of the total variability. As shown on fig. 4, CV1 reflects displacement of landmarks 4 and 5 toward each other. The CV1 axis (66,93% of total variability) exhibited a variation from a wider wing pointed apically (Poecilobothrus species) to an elongated wing with more obtuse apex (Dolichopus and Gymnopternus species).
The second canonical variate (CV2) accounted for 17,15% of the variance, was associated with displacements of landmarks 2 and 3 toward the posterior margin of the wing, and can be described as a variation from a wider wing (Gymnopternus species) to a wing pointed apically (Sybistroma). CV2 clearly separated Sybistroma and Gymnopternus species.
CV1
Fig. 4. Scatter plot from the first two canonical variates of wing shape for the five Dolichopodinae genera with the associated shape changes
Table
Assignment of specimens to their aprori defined groups (genera)
1 2 3 4 5 6 % correct
1. Dolichopus 3458 72 233 12 1 0 91.58
2. Hercostomus 73 888 77 123 31 0 74.50
3. Gymnopternus 161 4 1027 0 0 0 85.99
4. Poecilobothrus 16 256 0 1357 0 0 83.30
5. Sybistroma 0 32 0 0 154 0 82.80
6. Sympycnus 0 0 0 0 0 74 100,00
Total 3708 1251 1337 1492 187 74 86.32
Rows: observed classification; columns - predicted classification
The scatter plot from CV1 and CV2 showed some overlap among studied genera (fig. 4); this fact was confirmed by the results of discriminant analysis (table). The percentage of correctly classified was only about 86%, which indicated wing shape is not a reliable predictor of intergeneric discrimination. Dolichopus was the best-assigned genus, while Hercostomus had the lowest accuracy.
Ten percent of Hercostomus specimens (namely, H. apollo and H. phoebus) resembled and were classified as Poecilobothrus. Overall 83,3% of Poecilobothrus specimens were correctly identified to genus; the rest of the specimens belonging mainly to P. chrysozygos were classified as Hercostomus. Misclassifications of Gymnopternus have mainly been caused by cases of classifying the Gymnopternus specimens as those of Dolichopus. These cases of misclassification were not confined to certain species and were likely caused by measurement error.
Identification of species
Nineteen species (Dolichopus meigeni, D. ungulatus, D. latelimbatus, D. lepidus, D. longitarsis, D. rezvorum, D. ringdahli, D. sabinus, Gymnopternus metallicus, Hercostomus convergens, H. germanicus, H. pterostichoides, H. rivulorum, Poecilobothrus chrysozygos, P. nobilitatus, P. regalis, P. varicoloris, Sybistroma crinipes and S. obscuripes) were the assigned with the highest precision rate, reaching 95±5% of accuracy. Gymnopternus assimilis
demonstrated the lowest assignment rate (0,00%); 40% of its specimens were assigned as Gymnopternus aerosus and 60% as G. celer. Another case of the lowest accuracy (0,00%) was demonstrated by Hercostomus daubichensis, the specimens of which were identified as Gymnopternus aerosus, G. celer, G. pseudoceler and G. ussurianus. Hercostomus fulvicaudis, H. vivax, and P. clarus also exhibit low percentages of correct assignments (from 23 to 35%). Otherwise the classification success on species level ranges from 42,85 to 89,58%.
The following species of the genus Dolichopus showed a trend to cluster in the shape space: D. cilifemoratus, D. claviger, D. rezvorum, D. migrans, D. plumitarsis, D. ptenopedilus, D. simius, D. linearis (PC1 from 0,075 to 0,010; PC2 from 0,02 to 0,05) (Fig. 3). These species showed a combination of both morphometric traits: displacement of dm-m toward the distal margin of the wing and a widened wing. This combination stands out in the subfamily, because otherwise displacement of dm-m toward the distal margin of the wing was combined with narrowing of wing (P. nobilitatus, H. appolo).
The position of Poecilobothrus species within the morphospace should also be the focus of attention. Although the genus was represented by relatively few species, its confidence ellipse occupied the widest space, because Poecilobothrus species were occupying an extreme position in relation to the axis PC1. P. varicoloris were placed in the bottom right corner (PC1 = 0,07) and had a shorter mj section; and P. comitialis were placed in the bottom left corner (PC1 = -0,1) and had elongated mr
Discussion
The first result of our study is the determination of taxonomic value of wing shape for dolichopodid flies. Analysis based on the shape data classified 86,32% of the specimens to the correct genera and about 84% of the specimens to the correct species. Therefore, our results show that geometric morphometric analysis has capability for discriminating taxa of dolichopodine flies, but wing shape as taxonomic character is of limited use and must be complemented by traditional morphological traits.
Broad overlap zone between Hercostomus and other genera, which has been exhibited by canonical variate analysis, was a major issue. This leads to a suggestion that there are important challenges in the finding of diagnostic characters that can be used to distinguish between Hercostomus and other dolichopodine genera or separation of Hercostomus by subgenera.
The results derived from wing shape analysis were not always consistent with standard Dolichopodinae taxonomy. Wing shape of H. phoebus and H. appolo is more similar to those presented in the genus Poecilobothrus, while they formally would not be included in this genus on the basis of cercus morphology (H. phoebus has strip-shaped cercus and H. appolo has oval ones) and the absence of dark spot above the notopleuron. In contrast, Poecilobothrus chrysozygos tended to cluster with Hercostomus species in the shape space.
With regard to these cases, two explanations are possible. On the one hand, such close similarity between the species with respect to wing shape may reflect close phylogenetic relationship between species. On the other hand, previous studies showed varying degrees of phylogenetic signal in wing shape. The data suggest that there were parallel trends indicating a presence of homoplasy. Therefore, distantly related species also may tend to cluster together in the shape space.
Natural selection acting on flight behavior may lead to similar wing shape in not closely related species. Such ecological drivers of wing evolution could include the following: microhabitat-, predator-, or prey selection [19]. It leads to suggestion that both closely related species and species that occupy the same microhabitat may tend to cluster in the shape space.
With respect to assign exemplars into their a priori species, some species were distinguished more precisely than others. It could be assumed that the reason for low percentages of correct classification was the small sample size (Gymnopternus assimilis, Hercostomus daubichensis); but other species represented by small number of individuals were correctly reassigned to their
respective species (Dolichopus meigeni - 100%, Hercostomus eugenii - 80%, H. rivulorum - 100%). It is likely that cases of environmental specialization result in clearly different wing shape (for instance, Dolichopus meigeni, Hercostomus eugenii, H. rivulorum).
The second notable result of our study was the establishment of evolution trends in wing shape in the subfamily Dolichopodinae. Overall shape changes primarily occurred by the displacement of: (a) 5, 6 and 7 landmarks, which determine the position of posterior crossvein dm-m and apices of M+4; (b) 3 and 4 landmarks, determining the position of the apices of R4+5 and M1+2; and (c) 1 landmark - the insert point of R2 with costa.
Comparison with outgroup taxa reveals that proximal position of posterior crossvein dm-m and a longer R} are plesiomorphic in the subfamily Dolichopodinae, so the wing shape of Sybistroma can be construed as the most plesiomorphic in the subfamily. Other dolichopodine genera are characterized by a more distal position of dm-m. This result is in good agreement with current phylogeny hypotheses.
Second trend within the subfamily is displacement of the apices of R4+5 and M1+2, resulting in the apex of wing exhibiting trend toward a more pointed shape. According to phylogenetic relationships of dolichopodid flies, wing pointed apically as a discrete character has evolved independently several times in the subfamily Dolichopodinae by different ways: (a) Sybistroma exhibited a shift of landmark 4 toward a more anterior position, which correlated with displacement of landmark 7 toward the distal margin of the wing; (b) wing shape of Hercostomus and Poecilobothrus was formed through displacements of landmark 3 toward the posterior margin of the wing; (c) several Dolichopus species showed a convergence of R4+5 and M1+2 combined with the displacements of their apical sections toward the anterior wing margin.
Ennos [20] identified three functional wing types. According to this classification, Dolichopodidae possess wings capable of ventral flexion. The wings of this type can be bent in the basal part, another flexion line occurs in distal part of the wing. A change in location of termius Rj toward the base of the wing causes shifting of ventral flexion line, while changes in the placements of the crossvein, cubitus and media forming a "false margin" results in displacement of distal flexion line. Generally, the closer posterior crossvein is to the wing margin and the closer terminus of R2 is to the wing base, the more accurately will they be able to control the wing shape.
Another trend that has been observed in course of the analysis is change in the wing width through displacements of landmark 5 along the Y axis, which was discovered in Poecilobothrus, Hercostomus and Dolichopus species. This trend toward broadening of the wing blade suggests that these species are better adapted to behaviors requiring maneuverability during flight (for greater predator escape ability for instance [21]) than the more plesiomorphic species.
R e f e r e n c e s
1. Grichanov I. Ya. A checklist of species of the family Dolichopodidae (Diptera) of the World arranged by alphabetic list of generic names. (http://grichanov.aiq.ru/Genera3.htm), 2018.
2. Brooks S. E. Systematics and phylogeny of Dolichopodinae (Diptera: Dolichopodidae) // Zootaxa. -2005. - Vol. 857. - P. 1-158.
3. Germann C., Pollet M., Wimmer C., Bernasconi M. V. Molecular data sheds light on the classification of long-leg flies (Diptera: Dolichopodidae) // Invertebrate Systematics. - 2011. - Vol. 25. - P. 303-321.
4. Pollet M. A critical note on the systematic position of Gymnopternus (Diptera: Dolichopodidae) // Studia dipterologica. - 2003. Vol. 10, №2. - P. 537-548.
5. Stackelberg A. A. 29. Dolichopodidae // Die Fliegen der Palaearktischen Region. - 1933. - Bd. 4, №5, Lief. 71. - S. 114-179.
6. Khaghaninia S., Gharajedaghi Y., Grichanov I. Ya. Study of the genera Hercostomus Loew, 1857 and Poecilobothrus Mik, 1878 (Diptera: Dolichopodidae) in Kandovan Valley with new records for Iran // Biharean biologist. - 2013. - Vol. 7, № 2. - P. 73-79.
7. Negrobov O. P. Family Dolichopodidae // Catalogue of Palaearctic Diptera. Volume 7. Dolichopodidae-Platypezidae. - Budapest: Akademiai Kiado, 1991. - P. 11-139.
8. Bernasconi M. V., Pollet M., Ward P. I. Molecular systematic of Dolichopodidae (Diptera) inferred from COI and 12S rDNA gene sequences based on European exemplars // Invertebrate Systematics. - 2007.
- Vol. 21. - P. 453-470.
9. Bickel D. J. 49. Dolichopodidae (long-legged flies) // Brown B. V., Borkent A., Cumming J. M., Wood D. M., Woodley N. E. & Zumbado M. A., eds, Manual of Central American Diptera. Volume 1. -Ottawa: NRC Research Press, 2009. - P. 671-694.
10. Gelbic I., Olejnicek J. Ecology of Dolichopodidae (Diptera) in a wetland habitat and their potential role as bioindicators // Central European Journal of Biology. - 2011. - Vol. 8. - P. 118-129.
11. Torres A., Miranda-Esquivel D. R. Wing shape variation in the taxonomic recognition of species of Diachlorus Osten-Sacken (Diptera: Tabanidae) from Colombia // Neotropical Entomology. - 2015.
- Vol. 45, №2. - P. 180-191.
12. Schutze M. K., Jessup A., Clarke A. R. Wing shape as a potential discriminator of morphologically similar pest taxa within the Bactocera dorsalis species complex (Diptera: Tephritidae) // Bulletin of Entomological Research. - 2012. - Vol. 102. - P. 103-111.
13. Dvorak V., Aytekin A. M., Alten B., Skarupova S., Vatypka J., Volf P., A comparison of the intraspecific variability of Phlebotomus sergenti Parrot, 1917 (Diptera: Psychodidae) // Journal of Vector Ecology. - 2006. - Vol. 31, №2. - P. 229-238.
14. Sontigun N., Sukontason K. L., Zajac B. K., Zehner R., Sukontason K., Wannasan A., Amendt J. Wing morphometrics as a tool in species identification of forensically important blow flies of Thailand // Parasites & Vectors. - 2017. - Vol. 10. - P. 1-14.
15. Pepinelli M., Spironello M., Currie D. C., Geometric morphometrics as a tool for interpreting evolutionary transitions in the black fly wing (Diptera: Simuliidae) // Zoological Journal of the Linnean Society. - 2013. - Vol. 169. - P. 377-388.
16. Ulrich H. Zur systematischen Gliederung der Dolichopodiden (Diptera) // Bonner Zoologische Beitrage. - 1981. - Vol. 31. - P. 385-402.
17. Lim G. S., Hwang W. S., Kutty S. N., Meier R., Grootaert P. Mitochondrial and nuclear markers support the monophyly of Dolichopodidae and suggest a rapid origin of the subfamilies (Diptera: Empidoidea) // Systematic Entomology. - 2010. - Vol. 35. - P. 59-70.
18. Bookstein F. L. Morphometric tools for landmark data: geometry and biology. - New York: Cambridge University Press, 1991. - P. 435.
19. Chazot N., Panara S., Zilbermann N., Blandin P., Poul Y. L., Cornette R., Elias M., Debat V. Morpho morphometrics: shared ancestry and selection drive the evolution of wing size and shape in Morpho butterflies // Evolution. - 2015. - Vol. 70, №1. - P. 181-194.
20. Ennos A. R. Comparative functional morphology of the wings of Diptera // Zoological journal of the Linnean Society. - 1989. - Vol. 96. - P. 27-47.
21. Combes S. A., Daniel T. L. Shape, flapping and flexion: wing and fin design for forward flight // The Journal of Experimental Biology. - 2001. - Vol. 204. - P. 2073-2085.
