The Structure of the Endophallus Is a New Promising Feature and a Key to Study of Taxonomy of the Subgenus Metallotimarcha of the Genus Timarcha (Coleoptera, Chrysomelidae) in the Caucasus

Simple Summary Knowing the correct names of animal species is especially important nowadays; information is posted on the internet, and is available by using keywords. To establish the valid species and subspecies names, it is necessary to research a large amount of material and literature. Of particular importance is the study of type specimens—that is, those specimens for which the names were first established. In the subgenus Metallotimarcha, of the leaf-beetle genus Timarcha, the external structure of beetles and the shape of the male aedeagus do not allow precise identification of the species. The search for new characters has led to the need to study the internal structure of the male aedeagus. As a result, clear species boundaries of T. metallica, T. corinthia, and T. hummelii have been found. It was established that T. armeniaca and T. hummelii starcki are the synonyms of T. hummelii hummelii. Abstract A comparative morphological study of the members of the subgenus Metallotimarcha revealed that Caucasian Timarcha hummelii is a separate species, which differs from European members of the subgenus (T. metallica, T. corinthia, T. gibba) by the internal structure of male aedeagus, namely, manubrium of the endophallus. Morphology of manubrium is described for all species of the subgenus. External characters and the shape of male aedeagus do not permit separating the members of the subgenus distinctly. Examination of both external characters and endophallus structure in the specimens from the Caucasus revealed that T. armeniaca and T. hummelii starcki are the synonyms of T. hummelii hummelii, but not separate species or subspecies. Neotypes of T. hummelii and T. armeniaca are designated. Syntypes of T. hummelii starcki are examined. The key to the species for the subgenus Metallotimarcha is compiled.


Introduction
A new genus-group name Timarcha was originally published by Samouelle [1]. The later work by Latreille [2] is sometimes cited as the original description [3,4]. Samouelle [1] proposed the name Timarcha without a description, but with the available specific name Chrysomela tenebricosa Linnaeus clearly included under it. Thus, the name Timarcha Samouelle, 1819 is available [5] (Art. 12.2.5).
The subgenus Metallotimarcha Motschulsky, 1860 of the genus Timarcha Samouelle 1819 was originally described by Motschulsky [6] as a genus with the type species Chrysomela metallica Laicharting, 1781 by the original designation. The type species, Timarcha metallica, originally published as Chrysomela metallica Laicharting, 1781, is a primary junior homonym of Chrysomela metallica DeGeer, 1778 [7,8]. The two species have been included in different genera for a long time. Chrysomela metallica Laicharting was combined with the generic Central and southwestern Georgia, Northeastern Turkey 27 29 56 North and East Georgia, Armenia, Azerbaijan 18 17 35 Transcaucasia total 45 46 126 Western Europe 26 26 52 Balkans 6 9 15 Regions are grouped according to the territories from which the nominal taxa of the subgenus were described. Detailed information on all examined specimens is included in the Supplementary Materials.

Methods of Examination of Male Endophallus
First, the endophallus structures in all available males was studied. As rightly noted by Daccordi et al. [27], if a preparation of the apodemes is performed, the internal sac of the endophallus is destroyed. Conversely, if the internal sac of the endophallus is everted, the amodemes cannot be fully observed. Attention was placed on the preparation and study of apodemes as more sclerotized and permanent parts that can be prepared more easily to obtain comparable results. In view of the loss of the type specimens of T. metallica, the endophallus of the males from the type locality was studied.
Before dissection, the males were found by the dilated tarsomeres 1-3 and the peculiar structure of the last ventrite.

1.
The male specimen was put in water with a drop of detergent for 12-24 h at 20 • C for softing. 2.

4.
Aedeagus was washed in a large amount of water and the apodemes were extracted with a pin hooked at the end through the basal opening of the aedeagus.

5.
Apodemes were examined under a stereomicroscope. Then the aedeagus and apodemes were placed in glycerin in a Genitalia Micro Vial GVP 016 microtube pinned under a specimen.

Methods of Examination of External Morphology and Male Aedeagus
After the examination of endophallus, 12 more morphological characters were studied, including the characters that were used by the authors of the original descriptions of taxa and keys for the species of the subgenus Metallotimarcha.

Methods of Examination of External Morphology and Male Aedeagus
After the examination of endophallus, 12 more morphological characters were studied, including the characters that were used by the authors of the original descriptions of taxa and keys for the species of the subgenus Metallotimarcha.
Objective methods for recording color morphs, such as colorimetry or spectrometry, may provide better results and may be the subject of future research. However, at present, they are rarely used in taxonomic practice. The simple designation of colors with words prevails in the taxonomic literature and allows comparison of the results of different studies. In the case of the group under consideration, color morph is an additional feature that does not give 100% difference.

Criteria of Subspecies
The author follows the classical rule [40,41], which was recently (successfully) tested by Bieńkowski, Orlova-Bienkowskaja [42]: 97% of specimens in one sample should be separable from 97% of specimens in the other sample, to qualify these samples as representing different subspecies. Amadon [40] showed that this rule is fulfilled for a metric character if the following inequalities are true: M1 is the mean value of the variable in the first population, M2 is the mean value of the variable in the second population, σ1 is the standard deviation in the first population, and σ2 is the standard deviation in the second population [40]. Subspecies must be defined on diagnosability, not on average differences [43]. We use the classical statistical method for the distinguishing of the subspecies because this is the only method appropriate for distinguishing of the subspecies currently used in zoology, mainly in ornithology and theriology, e.g., [43][44][45][46][47][48]. Unfortunately, the overwhelming majority of insect subspecies are currently described or revised without any statistical treatment, e.g., [14,[49][50][51][52].   [23] in the second part of "Fauna Entomologica Trans-caucasica" described two species T. hummelii and T. armeniaca. Type localities were not indicated in the original descriptions. However, in the preface to the first part of "Fauna Entomologica Trans-caucasica", Faldermann [53] noted that the work was based on the collections of the expeditions by É.P. Ménétries and A.I. Szovitz in the Transcaucasia in 1829−1830.
Faldermann [23] did not indicate the number of specimens studied, but based on their sizes without the limits of variability, it can be assumed that each taxon was described based on the one specimen. The beetles collected during the expeditions by É.P. Ménétries and A.I. Szovitz were deposited mainly in ZIN, some specimens could have been donated to V.I. Motschulsky [54].
After Horn, Kahle [55], types of the taxa described by F. Faldermann, were donated to G.V. Mniszech (that means they can be in MNHN), and deposited in ZIN and ZMMU.
The first historical label refers to the original description, the second part of "Fauna entomologica Trans-caucasica" [23]. The page number is written incorrectly, the description of T. hummelii begins with p. 351. The general appearance of the label and the handwriting differ sharply from the original Faldermann labels [55,56]. The G.V. Mniszech collection included the specimens from different collections, donated by different persons, and not only from the Ménétries or Faldermann collections. Due to the absence of original or type labels, labels indicating the type locality, or labels indicating the specimen from Ménétries or the Faldermann collection, it is impossible to definitely attribute this specimen as a type. The specimen in MNHN is a female (according to the structure of the last abdominal sternite and narrow tarsomeres 1-3). Faldermann [23] described clearly a different specimen when he noted for T. hummelii: "tarsis <...> valde dilatatis" (=tarsis very wide).
Recent "holotype" labels under the specimen in MNHN are incorrect. The holotype was not designated in the original publication by Faldermann [23]. In this case, all type specimens should be considered as syntypes [5] (Art. 73.1.3).
There were no specimens labelled T. armeniaca in MNHN (the curator, A. Mantilleri, personal communication).
In ZIN, the type specimens of T. hummelii and T. armeniaca were not found. The Motschulsky collection (ZMMU) includes a male labelled "Timarcha hummeli" and a female labelled "Timarcha armeniaca". The male T. hummelii bears the characteristic Motschulsky collection label with the words "Conf. Persiae" (=at the border of Persia, Iran) (Figures 3a and 4b).
The borders of Russia and Persia, according to the Turkmanchay treaty of 1828, were established in Transcaucasia along the southern borders of the Erivan and Nakhichevan khanates and Azerbaijan [57]. E.P. Ménétries approached this border in Lankaran [54], and A.I. Szovitz crossed the border in Karabakh [58]. Finally, V.I. Motschulsky in 1837 accompanied the Iranian embassy on its way from Russia to the border [59].
Thus, a specimen of T. hummelii in the Motschulsky collection could have been collected by Ménétries in Lankaran or by Szovitz in Karabakh and, therefore, could belong to the syntypes, or could have been collected later by Motschulsky near the border of Azerbaijan or Nakhichevan with Iran.
The female T. armeniaca bears old label "Achalzik Abas Tuma" (=Akhaltsikhe, Abastuman, villages in S.-W. Georgia) and the characteristic Motschulsky collection label with the locality "Armenia" (Figures 3b and 4c). Ménétries did not visit southwestern Georgia. Szovitz returned to Tbilisi at the end of the expedition and investigated the flora of Mingrelia and Imeretia. Motschulsky visited Akhaltsikhe in the spring of 1835. Thus, this specimen could have been collected by Szovitz (and belong to syntypes), or it could have been collected later by Motschulsky. The indication "Armenia" probably appeared on the collection label as a suggested type locality, based on the specific name "armeniaca". collected later by Motschulsky. The indication "Armenia" probably appeared on the collection label as a suggested type locality, based on the specific name "armeniaca".  Both specimens in the Motschulsky collection (ZMMU), T. hummelii and T. armeniaca, have no type labels or other type markings. They cannot be definitely attributed with the original type specimens. To study the systematic position of these taxa, it is necessary to distinguish neotypes. These specimens are suitable for this purpose. They were definitely collected in the Transcaucasia, the region, by the fauna of which Faldermann [23] worked. These specimens were obviously studied by Motschulsky [6]; they belong to the subgenus Metallotimarcha established by him, they correspond to the recent interpretation of T. hummelii and the interpretation of T. armeniaca as its synonym, thereby ensuring the stability of the nomenclature.  Both specimens in the Motschulsky collection (ZMMU), T. hummelii and T. armeniaca, have no type labels or other type markings. They cannot be definitely attributed with the original type specimens. To study the systematic position of these taxa, it is necessary to distinguish neotypes. These specimens are suitable for this purpose. They were definitely collected in the Transcaucasia, the region, by the fauna of which Faldermann [23] worked. These specimens were obviously studied by Motschulsky [6]; they belong to the subgenus Metallotimarcha established by him, they correspond to the recent interpretation of T. hummelii and the interpretation of T. armeniaca as its synonym, thereby ensuring the stability of the nomenclature.
The female from MNHN is not suitable for the designation of neotype of T. hummelii for two reasons: (1) for the neotype of the valid taxon, a male is preferable to the female in a group where the main diagnostic features are the structure of the male endophallus.
(2) A more detailed type locality is preferably for the taxon, which may contain subspecies, for more accurate understanding of the nominotypical subspecies.

Timarcha hummelii starcki
Bechyné [26] described a subspecies starcki from the "Western Caucasus" based on an unspecified number of specimens, without a designation of the holotype. Subsequent authors erroneously indicated the Eastern Caucasus as the range of this subspecies [3] or included Armenia in it [4]. In NMB, there are two syntypes of the subspecies T. hummelii starcki collected by the famous Russian expert on bark beetles V.N. Starck, with the label The female from MNHN is not suitable for the designation of neotype of T. hummelii for two reasons: (1) for the neotype of the valid taxon, a male is preferable to the female in a group where the main diagnostic features are the structure of the male endophallus.
(2) A more detailed type locality is preferably for the taxon, which may contain subspecies, for more accurate understanding of the nominotypical subspecies.

Timarcha hummelii starcki
Bechyné [26] described a subspecies starcki from the "Western Caucasus" based on an unspecified number of specimens, without a designation of the holotype. Subsequent authors erroneously indicated the Eastern Caucasus as the range of this subspecies [3] or included Armenia in it [4]. In NMB, there are two syntypes of the subspecies T. hummelii starcki collected by the famous Russian expert on bark beetles V.N. Starck, with the label "Cauc. Occid. Regio maritima" (=Western Caucasus, seashore region) (Figures 3d and 4d). The Western Caucasus is a part of the Greater Caucasus mountain system, located to the west of the meridional line passing through Mountain Elbrus [60]. Therefore, the type locality is the southern foothills of the Western Caucasus near the Black Sea coast according to [5] (Art. 76A.1.1.). The author of the present study does not designate a lectotype. The designation of a lectotype is justified when there is reason to believe that a type series may include more than one taxon. As a result of the present study of syntypes, it was established that both of them belong to the same taxon. In this case, the designation of the lectotype is optional.

Timarcha metallica
This species was described based on an unspecified number of specimens from Tyrol [8]. The J.N. Laicharting collection was deposited in Tiroler Landesmuseen Ferdinandeum, Innsbruck. This collection was lost many years ago [61]. The type of T. metallica is missing from the museum (the curator, A. Eckelt, personal communication). There is no need to designate the neotype, since the interpretation of this taxon is clear and unambiguous in recent taxonomical literature [3,13,27,62,63]. If it is needed, for some reason, then a specimen from Tyrol should be designated and it should be placed in the Tiroler Landesmuseen, where the Laicharting collection was originally deposited. The topotypes from Tyrol were examined.

Timarcha hummelii
All studied males from the Caucasus, including T. hummelii neotype (Figure 1b,c), T. hummelii starcki and T. armeniaca topotypes (Figure 1d), have a similar type of apodemes structure. There is insignificant variability in the width of the base of the manubrium, but it has no geographical character, since it is sometimes present in specimens collected in one locality at the same time.
The manubrium is 1.22-1.42 mm long, well sclerotized, with rudimental wings; the manubrium is narrow, elongate, hardly curved dorso-ventrally, parallel-sided in apical 3/4, broadest in basal 1 ⁄4, with a semi-transparent medial part at the base, with a narrow furrow consisting of flagellum along with the entire length, ending with two transparent narrow apical lobes. Variability consists of more or less broadened basal 1 ⁄4 of manubrium. Flagellum is long, narrow, and tube-shaped.
Manubrium (Figure 1e) is 0.74-0.86 mm long, well sclerotized, without distinct wings, but with paired basal lobes. Manubrium is straight in the lateral view, broadest and quadrangular (in dorsal view) in basal 1 ⁄2, with two triangular basal lobes curved ventrally, narrowest after mid-length, and slightly broadened, bearing irregular denticles along the outer margin at the apex, with narrow furrow consisting of flagellum along the entire length. Flagellum is long, narrow, and tube-shaped.

Timarcha gibba
Specimens from Europe, collected near the type locality of T. gibba (mountains in the environs of Trieste, partly in Slovenia), and corresponding to the original description [65] and subsequent interpretation of this taxon [3,13,16,17,64], do not differ in the structure of apodemes from T. metallica.
Manubrium is 0.85 mm long. Flagellum is long, narrow, and tube-shaped.

Timarcha corinthia
Timarcha corinthia was originally described from Dalmitia (=Croatia + part of Montenegro). Specimens from Serbia, Bosnia-Herzegovina, Montenegro, corresponded to the original description [10] and subsequent interpretation of T. corinthia [3,13,16,17] have apodemes very different from T. hummelii and T. metallica in the presence of wings at the base of manubrium.
Manubrium (Figure 1f) is 1.71-1.94 mm long, well sclerotized, with long paired basal wings; manubrium is narrow, elongate, hardly curved dorso-ventrally, broadest at the base, and gradually narrowed from the base to apex, with narrow furrow consisting of flagellum along the entire length. Flagellum is long, narrow, and tube-shaped.

Results of the Quantitative Comparison of Samples from Different Regions
Males were identified by the genitalia (manubrium structure). Females, collected at the same time with certain males or in a region from which only one species is known, were identified accordingly. Samples from populations are grouped as follows. Regions 1, 2, and 3 represent the area of T. hummelii starcki, 4-the area of T. armeniaca, 5-the area of T. hummelii hummelii, 6-the total area of T. armeniaca + T. hummelii hummelii, 7-the area of T. metallica, 8-the area of T. corinthia.
Comparison of T. corinthia, T. metallica, and T. hummelii. It was found (Tables 2 and 3) that T. corinthia is mostly larger than T. metallica, but there is no clear interspecific difference (hiatus) by the length of all specimens, as well as, separately, males and females. Timarcha hummelii is mostly larger than T. metallica, but there is no species difference too.  Bechyné [26] noted that females of T. hummelii starcki are larger than those of T. hummelii hummelii. According to the available specimens, the largest female from the Western Caucasus is 12.24 mm long, and the largest female from Transcaucasia not smaller, but even slightly larger, 12.46 mm long. The populations from the WesternCaucasus and Transcaucasia do not correspond to Amadon's criteria [40] of the subspecies by this character.

Size of the Pronotum (a): Pronotal Length (in Dorsal View)/Elytral Length (in Lateral View)
According to Weise [16], Medvedev, Shapiro [20], the pronotum is "large" in T. hummelii, and "small" in T. metallica. However, the authors did not indicate which parameters were taken into account. Two parameters: relative length (present feature) and relative width (next feature) of the pronotum were selected in the present work (Table 4).   (Table 5). According to Faldermann [23], the maximal width of the pronotum is at mid-length in T. hummelii, and before the mid-length in T. armeniaca. According to [10], the maximal width of the pronotum is at base in T. gibba, and before the mid-length in T. metallica. According to Weise [16], the maximal width of the pronotum is at base in T. gibba, pronotum narrowed anteriorly (slightly more) and posteriorly in T. corinthia, T. metallica, and T. hummelii. According to Marseul [17], the maximal width of the pronotum is at midlength in T. armeniaca, and before the mid-length in T. hummelii. According to Medvedev, Shapiro [20], the maximal width of the pronotum is almost at the anterior margin in T. hummelii, and before the mid-length in T. metallica. According to Warchałowski [3], the maximal width of the pronotum is at the base in T. gibba, at mid-length in T. corinthia, and before mid-length in T. metallica and T. hummelii (Table 6). Table 6. Location of the maximal width of the pronotum: distance from the level of the front corners to the level of the greatest width of the pronotum/total length of the pronotum (both in dorsal view). M-mean value, σ-standard deviations, SE-standard error. Comparison of T. corinthia, T. metallica, and T. hummelii Timarcha corinthia has the greatest width of the pronotum, on average, closer to the apex than T. metallica, but there is no clear interspecific difference (hiatus). In T. hummelii, the greatest width of the pronotum is on average closer to the apex than in T. metallica, but there is no clear interspecific difference (hiatus). Therefore, |M6 − M3| < 3.24σ6 + 0.68σ3 and |M9 − M3| < 3.24σ3 + 0.68σ6. The difference does not reach the level of the subspecies.

Emargination of Pronotal Lateral Side before Base (Present or Absent)
According to Faldermann [23], the lateral side is slightly emarginate before the base in T. armeniaca, and almost without emargination in T. hummelii. According to Marseul [17], the lateral side is distinctly emarginate before the base in T. corinthia and T. armeniaca, slightly emarginate in T. hummelii, and without emargination in T. gibba and T. metallica. According to Medvedev, Shapiro [20], the lateral side is slightly emarginate in T. hummelii, and more or less rounded in T. metallica (Table 7). Comparison of T. corinthia, T. metallica, and T. hummelii. Specimens of T. corinthia has emargination more often than T. metallica and T. hummelii, but there is no clear interspecific difference (hiatus).
Intraspecific variability of T. hummelii. Comparison of regions 4 and 5. Populations from these regions hardly differ in this character.
Comparison of regions 3 and 6. In individuals from the Western Caucasus, the emargination presents more often than in individuals from the Transcaucasia, but the difference does not reach the level of subspecies.

Dorsal Color
Different authors could designate the same colors in different ways. In the present work, a spectrum of colors that reflects variability is adopted. According to Faldermann [23], the dorsal color is purple copper with an elytra dark copper in T. hummelii, and copper green with an elytra dark greenish copper in T. armeniaca. According to Weise [16], dorsum is brassy in T. corinthia, brown with strong brass shine in T. metallica, brown, or violet, or greenish, with copper shine in T. hummelii, piceous with violet or blue shine in T. gibba. According to Marseul [17], the dorsal color is golden bronze in T. corinthia, black in T. gibba, and bronze brown in T. metallica. According to Medvedev, Shapiro [20], the dorsal color is rusty red or purple in T. hummelii, and darker, bronze or coppery in T. metallica. According to Warchałowski [3], the dorsal color is greenish bronze, copper, blue, or violet in T. corinthia, and almost black with blue sheen in T. metallica (Table 8). Comparison of T. corinthia, T. metallica, and T. hummelii.
In T. corinthia and T. metallica, colors 5 and 6 are sharply predominant, in T. hummelii, they are not found, but colors 1 and 4 are predominant. Due to the presence of color types 1-4 in the first two species, there is no clear interspecific difference (hiatus).
Intraspecific variability of T. hummelii. Comparison of regions 4 and 5.
Populations from these regions are very similar in color types. The difference does not reach the level of subspecies.
Comparison of regions 3 and 6.
Populations from the Western Caucasus and from the Transcaucasia are very similar in color types. The difference between them does not reach the level of subspecies.

Color of Femora
Most authors described the coloration of the legs in general. However, since different parts of the leg can be colored differently, the coloring of the femora (present feature) and tarsi (the next feature) were considered separately in the present work.
According to Faldermann [23], legs are brown, almost bronze, with tarsi light in T. hummelii, and legs are pitch-brown, slightly copper, with tarsi brown in T. armeniaca. According to Weise [16], legs are violet in T. corinthia; they are more or less red-brown in T. hummelii and T. metallica. According to Marseul [17], legs are violet in T. corinthia, and blue-black in T. gibba. According to Warchałowski [3,13], legs are red-brown or red in T. metallica and T. hummelii, and black or almost black with metallic shine in T. corinthia (Table 9). Comparison of T. corinthia, T. metallica, and T. hummelii. Timarcha corinthia differs sharply from T. metallica and T. hummelii in the predominance of color 3, but there is no clear interspecific difference (hiatus). In T. hummelii and T. metallica, color 2 is predominant; there is no clear interspecific difference (hiatus).
Intraspecific variability of T. hummelii. Comparison of regions 4 and 5: populations from these regions des not differ by the color.
Comparison of regions 3 and 6. The proportion of color 2 is slightly higher in region 9 than in region 3, but the difference does not reach the level of subspecies.

Color of Tarsi
Color of Tarsi (Table 10). Comparison of T. corinthia, T. metallica, and T. hummelii. Timarcha corinthia sharply differs from T. metallica and T. hummelii in the presence of color 3 in all studied individuals, but due to the presence of this color in a small proportion in T. hummelii and T. metallica, there is no clear interspecific difference (hiatus).

Comparison of regions 4 and 5.
Populations from these regions differ slightly. The difference does not reach the level of subspecies.
Comparison of regions 3 and 6.
Populations from these regions differ slightly. The difference does not reach the level of subspecies. According to Faldermann [23], elytral punctures are large in T. hummelii, and coarse in T. armeniaca. According to Weise [16], elytra are strongly punctate in T. corinthia, and, rather densely, more or less strongly punctate in T. metallica. According to Marseul [17], elytral punctures are more dense and coarse in T. gibba, and more sparse and small in T. metallica. According to Bechyné [26], elytral punctures are denser in T. hummelii starcki than in T. hummelii hummelii. According to Warchałowski [3], elytral punctures are dense in T. hummelii starcki, and moderately dense in T. hummelii hummelii (Table 11).
During the present study, it was found that the size of the puncture is a more variable parameter than the density. Comparison of T. corinthia, T. metallica, and T. hummelii. Timarcha corinthia sharply differs from T. metallica and T. hummelii due to the predominance of puncture state 6; due to the presence of this state in a small proportion in T. hummelii and T. metallica, there is no clear interspecific difference (hiatus).
Comparison of regions 4 and 5.
Populations from these regions slightly differ; the difference does not reach the level of the subspecies.
Comparison of regions 3 and 6. Populations from region 3 are distinguished by the predominance of punctation state 1, which is rare in region 9, but the difference does not reach the level of the subspecies.

Border at Upper Margin of Elytral Epipleura near Base (Present, Absent)
According to Marseul [17], the border is present in T. corinthia, T. gibba, T. metallica, and absent in T. armeniaca and T. hummelii (Table 12). Comparison of T. corinthia, T. metallica, and T. hummelii. Timarcha corinthia and T. metallica differ from T. hummelii by the presence of a border in most specimens, but there is no clear interspecific difference (hiatus).

Comparison of regions 4 and 5.
Populations from these regions does not differ. Comparison of regions 3 and 6. About a third of individuals from region 3 have borders, and no individuals from region 6 have borders. Thus, the difference does not reach the level of subspecies.
Shining males predominate in all species, shining females-in T. corinthia and T. metallica, but there is no clear interspecific difference (hiatus).
Intraspecific variability of T. hummelii. Comparison of regions 4 and 5. Region 5 has more dull females and males than region 4, but the difference does not reach the level of subspecies.
Comparison of regions 3 and 6. Region 6 has more dull females and males than region 3, but the difference does not reach the level of subspecies.

Key to Species (T. gibba Is Not Included There Because Its Taxonomical Position Is Unclear)
1. Species from Western Europe. Border at upper margin of elytral epipleura near the base mostly present...2.
-Species from the Caucasus and Asia Minor. Border at upper margin of elytral epipleura near the base is usually absent, rarely present. Male genitalia: manubrium 1.22-1.42 mm long, without distinct wings, narrow, elongate. Body 6.86-12.46 mm long. Dorsum usually violet or golden coppery; femora usually piceous, rarely rufous or black, tarsi usually rufous or piceous. The apex of aedeagus is mostly evenly curved...T. hummelii.
2. Male genitalia: manubrium 1.71-1.94 mm long, with long paired basal wings, narrow, elongate, broadest at the base, and gradually narrowed from the base to apex. Body 9.08-12.77 mm long. The dorsum is usually bronze, the femora is mostly black, and the tarsi black. The apex of aedeagus is evenly curved...T. corinthia.
-Male genitalia: manubrium 0.74-0.86 mm long, without distinct wings, broadest and quadrangular (in dorsal view) in basal 1 ⁄2. Body 6.86-10.50 mm long. The dorsum is usually bronze or blackish bronze, the femora is usually piceous, rarely rufous or black, the tarsi is usually piceous or black. The apex of aedeagus mostly recurved dorsally...T. metallica.

General Discussion and Conclusions
The structure of the male endophallus makes it possible to clearly identify T. hummelii, T. metallica, and T. corinthia. Most of the external features and the external structures of the aedeagus, including those indicated in the literature, do not allow distinguishing between the species of the subgenus Metallotimarcha. Additional features typical for most specimens are given below. Timarcha corinthia and T. hummelii are mostly larger than T. metallica. In T. corinthia and T. metallica, the body colors are mostly bronze and blackish bronze, while violet and golden coppery colors are predominant in T. hummelii. Timarcha corinthia differs sharply from T. metallica and T. hummelii in the predominance of black in the color of the femora, while the femora are mostly piceous in T. hummelii and T. metallica. Timarcha corinthia sharply differs from T. metallica and T. hummelii in the presence of black tarsi in all studied individuals, while this color is present in a small proportion in T. hummelii and T. metallica. Timarcha corinthia sharply differs from T. metallica and T. hummelii due to predominance of large elytral punctures, while this state of the character is present in a small proportion in T. hummelii and T. metallica. Timarcha corinthia and T. metallica differ from T. hummelii by the presence of a border at the upper margin of elytral epipleura near the base in most specimens.
We can also take into account the geographical feature: T. metallica, T. corinthia, and T. gibba inhabit Western Europe, and T. hummelii occur in the Caucasus and Asia Minor ( Figure 5). tarsi in all studied individuals, while this color is present in a small proportion in T. hummelii and T. metallica. Timarcha corinthia sharply differs from T. metallica and T. hummelii due to predominance of large elytral punctures, while this state of the character is present in a small proportion in T. hummelii and T. metallica. Timarcha corinthia and T. metallica differ from T. hummelii by the presence of a border at the upper margin of elytral epipleura near the base in most specimens.
We can also take into account the geographical feature: T. metallica, T. corinthia, and T. gibba inhabit Western Europe, and T. hummelii occur in the Caucasus and Asia Minor ( Figure 5). Timarcha gibba does not differ from T. metallica in the structure of endophallus. Taxonomical status of T. gibba has been questioned [66]. For a final decision, it is necessary to study additional material.
The manubrium of T. corinthia is very different from those in other Metallotimarcha members in the presence of wings. This feature is similar to that of Timarcha s. str., but the pronotum is not bordered, as typical of the subgenus Metallotimarcha. The width and shape of the mesosternum is mentioned as a distinctive feature of the subgenus Metallotimarcha [3,10,11,13,14]. According to my results, it does not clearly distinguish between the subgenera.
The endophallus structure (shape of manubrium) in males from the Caucasus has individual variability, but does not give geographic variability. Comparison of individuals from the populations of the West Transcaucasia and East of Transcaucasia (typical areas of T. armeniaca, and T. hummelii, respectively) and comparisons of individuals from Timarcha gibba does not differ from T. metallica in the structure of endophallus. Taxonomical status of T. gibba has been questioned [66]. For a final decision, it is necessary to study additional material.
The manubrium of T. corinthia is very different from those in other Metallotimarcha members in the presence of wings. This feature is similar to that of Timarcha s. str., but the pronotum is not bordered, as typical of the subgenus Metallotimarcha. The width and shape of the mesosternum is mentioned as a distinctive feature of the subgenus Metallotimarcha [3,10,11,13,14]. According to my results, it does not clearly distinguish between the subgenera.
The endophallus structure (shape of manubrium) in males from the Caucasus has individual variability, but does not give geographic variability. Comparison of individuals from the populations of the West Transcaucasia and East of Transcaucasia (typical areas of T. armeniaca, and T. hummelii, respectively) and comparisons of individuals from the West Caucasus and Transcaucasia does not give a difference of the sub-specific range. Thus, a synonymy T. armeniaca = T. hummelii is confirmed, and a new synonymy: T. hummelii starcki = T. hummelii hummelii is established. The molecular genetic study of the subgenus Metallotimarcha may become the subject of further research and will allow testing the conclusions of this article at a new level of knowledge.