Myrmica Species Groups

The genus Myrmica is primarily a Holartic genus. These ants are a common element of the ant faunas of large parts of Europe and North America. Species identification can be difficult and frustrating. Radchenko and Elmes have been been steadily and consistently improving the taxonomy of the Palearctic species for more than 2 decades, with Seifert also making important contributions in this effort. Franceour has been studying the North American Myrmica for decades but little of his research has been finalized and published.

The following largely follows Radchenko and Elmes (2010).

Myrmica species groups
When considering the range of variation between species within a genus it is usually apparent that some species are very similar to each other and are quite dissimilar to other clusters of species. It is normal practice in taxonomic studies to reflect this either by formally separating the genus into subgenera, or by using the less formal concept of species groups (note that the International Code of the Zoological Nomenclature does not regulate the concept of species groups or species complexes). The problem with a subgeneric division of the genus Myrmica is that even in the case of the ritae-group, which appears to us to have the clearest claim as a subgenus, we have found species intermediate to other clear groups; the number of clear subgenera would be so small that the vast majority of species would remain unassigned. The greater flexibility of the species groups concept makes it possible to assign a majority of Myrmica species to a group, but even so, a number of taxa are so dissimilar to all others that they can not be satisfactorily assigned to any group, or can be considered as a mono-specific group.

Species groups. Ideally, as in any supraspecific taxon, the species groups should include phylogenetically related species. We have assigned species to groups using characters of all three castes. Species are distinguished from related ones by at least one morphological autapomorphy, but in contrast species groups are defined by features the species hold in common and they can only be distinguished from other groups by a whole complex of features, some of which may widely overlap. In some cases, especially when only a single caste is known (usually workers) homoplasies (independently evolved similarities) will place unrelated species in the same group. Recently, the molecular studies of Jansen et al. (2009, 2010) showed that those of our proposed species groups that they included in their study, appear to be monophyletic; however, the same studies also indicate that generally the species complexes suggested by us appear to be paraphyletic, i.e. artificial complexes based on morphological similarities that do not necessarily reflect close relatedness. Even so, the match between Jansen et al.'s molecular phylogeny and our species groups gives us confidence that when the “gaps” in their phylogeny are “filled in” most of our species groups will be validated. In the mean time, the species group names are useful "shorthand" when discussing the distribution and evolution of the genus.

In this monograph we recognize 17 groups of Old World Myrmica (number of species in parenthesis); in alphabetical order they are: the arnoldii (1), dshungarica (4), inezae (3), karavajevi (3), kurokii (2), laurae (1), lobicornis (22), luteola (2), myrmicoxena (1), pachei (14), ritae (21), rubra (4), rugosa (9), scabrinodis (26), schencki (11), smythiesii (5) and tibetana (3) groups.

Below, we list the species groups alphabetically and amplify the diagnostic characteristics of each group, together with any special notes or observations, and give a list of species assigned to the groups. In some groups it is possible to recognise several species complexes. Generally, we do not recognise species groups based on a single species, except for the arnoldii-, laurae- and myrmicoxena-groups; the first contains a most unusual free-living species that has morphological characters often associated with social parasites, and the latter pair contains poorly studied social parasites. This approach leaves ten free-living species (listed after the species groups data) that cannot be placed in any species group unambiguously.

arnoldii group
Myrmica arnoldii

Diagnosis Workers: frontal carinae merging with the rugae that extend to the occipital margin, they do not curve outwards and do not merge with rugae that surround antennal sockets; frons wide, but frontal lobes extended. Scape quite strongly curved at the base but not angled and with no trace of lobe or carina. Petiole and postpetiole with at least small ventral lobes. Spurs on the middle and hind tibia reduced to various extents. Males: antennae 12-segmented, scape short, SI1 < 0.40.

This monospecific group was established to contain a most unusual free-living species from South Siberia and Mongolia that has in all castes morphological features often associated with social parasites. At the moment it is unique worldwide, but possibly similar species could be found in Tibet and China when these regions are more fully studied.

dshungarica group
Myrmica dshungarica

Myrmica ferganensis

Myrmica juglandeti

Myrmica kryzhanovskii

Diagnosis Workers: frontal carinae merging with the rugae that extend to the occipital margin, they do not curve outwards and do not merge with rugae that surround antennal sockets; frontal lobes slightly curved, frons wide, and frontal lobes not extended. Scape very smoothly curved at the base, not angled and with no trace of lobe or carina. Anterior clypeal margin convex, often prominent, with no medial notch. Males: scape long, SI2 > 1.00.

The species of this group are superficially similar to those of the rubra-group, but differ from the latter by the shape of frontal carinae. The group includes four species, all of which are restricted to the Middle Asian mountains and NE Afghanistan.

inezae group
Myrmica inezae

Myrmica mixta

Myrmica rigatoi

Diagnosis Workers: scape long, subequal to head length, smoothly curved at the base, not angled and with no trace of lobe or carina; frontal lobes slightly curved, frons wide and frontal lobes not extended. Anterior clypeal margin is distinctly prominent with no medial notch. Petiole has a very long and thin peduncle and a quite short and high node; postpetiole is subglobular. Propodeal spines very long; propodeal lobes rounded, not pointed apically. Males are unknown.

The species of this group share several features with ritae-group species (e.g. long scape, petiole and propodeal spines, and a slender body), but well differ from them by their strongly prominent and not-notched anterior clypeal margin and not-pointed apically (more rounded) propodeal lobes. We place three species from the Himalaya and south-western China in this group.

karavajevi group
Myrmica kabylica

Myrmica karavajevi

Myrmica lemasnei

Diagnosis Workerless social parasite. Queens: frontal carinae merging with the rugae that extend to the occipital margin, they do not curve outwards and do not merge with rugae that surround antennal sockets; frons wide, frontal lobes not extended. Scape gradually but quite strongly curved at the base, not angled and with no trace of lobe or carina. Petiole and postpetiole have distinct ventral lobes. Spurs on the middle and hind tibia reduced to various extents. Males: antennae 12-segmented, scape relatively long, SI1 > 0.70.

We place three species distributed in Europe and Algeria in this group, which equates to the subgenus Symbiomyrma (sensu Seifert 2007).

Myrmica social parasites. Although, we consider that M. karavajevi, M. lemasnei and M. kabylica are not sufficiently different from Myrmica to be considered as a separate genus, we agree with Seifert (1993b) that they differ from all other Myrmica on a subtle combination of size and body sculpture and we suspect they have different life-history strategies compared to the other socially parasitic Myrmica. Therefore we consider them as a new and distinct species-group (sensu Radchenko 1994a; Radchenko and Elmes 2001a) within Myrmica — the karavajevi group. The features of the karavajevi-group are small sized queens, no workers, a subpetiolar flange, a wide postpetiole, generally hairy, reduced tibial spurs, an individual cubital cell and males with 12 jointed antenna.

kurokii group
Myrmica kozlovi

Myrmica kurokii

Diagnosis Workers: overall appearance is large and robust. The frontal carinae merge with the rugae that extend to the occipital margin, they do not curve outwards and do not merge with rugae that surround antennal sockets. Scape gradually but quite strongly curved at the base, not angled and with no trace of lobe or carina. Anterior clypeal margin rounded, without a medial notch. Petiole has a short peduncle and massive, quite high node, Head dorsum has a very dense but not coarse rugosity, surface between rugae densely punctated. Males: scape long, SRI> 0.90.

We placed to this group two species, distributed from Himalaya and Tibet to Japan. Quite probably it will be found that this group comprises several more species when the fauna of Tibet and China is better known.

laurae group
Myrmica laurae

Diagnosis Workerless social parasite. Queens: frontal carinae merging with the rugae that extend to the occipital margin, they do not curve outwards and do not merge with rugae that surround antennal sockets; frons wide, frontal lobes not extended. Scape quite strongly but gradually curved at the base with a horizontal lobe (the combination of a gradually curved scape with a lobe is unique among all known Old World Myrmica species). Petiole and postpetiole with ventral lobes. Spurs on the middle and hind tibia reduced to a variable extent. Very hairy species; eyes with conspicuous hairs, length of the longest hairs ≥ 0.035 mm. Males: antennae 12-segmented, scape long, SII> 0.75.

This group contains a single species found in the Italian peninsula.

lobicornis group
The species are distributed among the complexes as follows:

lobicornis-complex
Myrmica anatolica

Myrmica eidmanni

Myrmica forcipata

Myrmica jessensis

Myrmica kirghisorum

Myrmica lobicornis

Myrmica lobulicornis

Myrmica pisarskii

Myrmica saposhnikovi

Myrmica wesmaeli

Myrmica zojae

excelsa-complex
Myrmica excelsa

Myrmica tamarae

Myrmica transsibirica

kasczenkoi-complex
Myrmica kasczenkoi

Myrmica commarginata

Myrmica angulinodis

Myrmica displicentia

Myrmica kamtschatica

sulcinodis-complex
Myrmica sulcinodis

Myrmica ademonia

xavieri-complex
Myrmica xavieri

Diagnosis Workers: scape strongly curved at the base, often with vertical lobe or dent, but sometimes only angled or curved, without dent; the lobe or dent, if present, never forms shield-like plate along the short (vertical) surface of scape (foot of scape). Anterior clypeal margin widely rounded, with a notch medially (except for M. xavieri that has no medial notch). Males: scape relatively long, SI1 > 0.60.

The lobicornis-group is one of the three most diverse species groups of the Old World containing 22 species that, taken together, are distributed throughout the Palaearctic Region. We have divided this group into five species complexes, based on worker characters (see below).

Characteristic of the species complexes within lobicornis species group (based on workers):

-Anterior clypeal margin not-notched medially - xavieri-complex

-Anterior clypeal margin notched medially

-Lateral portion of clypeus raised into sharp ridge in front of antennal insertions, so that antennal sockets distinctly separated from c1ypeal surface - excelsa-complex

-Lateral portion of clypeus not raised into sharp ridge in front of antennal insertions, so that antennal sockets lay on the same level with clypeal surface.

-Scape at the base with vertical lobe or at least dent - lobicornis-complex

-Scape gradually curved or angled at the base, without vertical lobe or dent

-Alitrunk with very coarse, straight longitudinal rugae; propodeal spines long, species' means ESLI ≥ 04.0; petiole with short peduncle, its anterior surface almost straight, meets with dorsal surface at an almost right or faintly blunt angle, dorsal plate well developed, flattened, not inclined posteriorly - sulcinodis-complex

-Alitrunk with less coarse sinuous longitudinal rugosity; the same level with clypeal lobe or propodeal spines shorter, species' means ESLI ≤ 0.35; petiole of various shape, but never with well developed, flattened dorsal plate - kasczenkoi-complex

luteola group
Diagnosis Workers: frontal carinae curved outwards to merge with the rugae that surround antennal sockets; frontal lobes slightly curved, frons wide, and frontal lobes not extended. Scape gradually though quite strongly curved at the base, not angled, with no trace of lobe or carina. Large and very hairy species. Males: scape short, SI1 < 0.50.

We place two species to this group, both are quite unusual compared with all other Myrmica species. The first is found in Eastern Asia, including Japan, while the second is known only from Taiwan:

Myrmica luteola

Myrmica mirabilis

myrmicoxena group
Diagnosis Workerless social parasite. Queens: frontal carinae merging with the rugae that extend to the occipital margin, they do not curve outwards and do not merge with rugae that surround antennal sockets; frons wide, frontal lobes not extended. Scape gradually while quite strongly curved at the base, not angled, without any trace of lobe or carina. Petiole and postpetiole with ventral lobes. Spurs on the middle and hind tibia reduced in various extents. Males: antennae 13-segmented, scape short, SI1 < 0.40.

This group contains a single species.

Myrmica myrmicoxena

pachei group
Myrmica heterorhytida

Myrmica hlavaci

Myrmica multiplex

Myrmica pachei

Myrmica phalacra

Myrmica pleiorhytida

Myrmica polyglypta

Myrmica sculptiventris

Myrmica schulzi

Myrmica taibaiensis

Myrmica varisculpta

Myrmica villosa

Myrmica weii

Myrmica yunnanensis

Diagnosis Workers: alitrunk dorsum at least partly with transverse rugosity. Scape gradually though distinctly curved at the base, not angled, with no trace of lobe or carina. Anterior clypeal margin rounded or slightly prominent with no medial notch. Petiole with a relatively short peduncle. Males: scape long, SI1 > 0.90 (but males of only M. pachei are known).

Previously we placed in this group only Himalayan species that we believed to be rare and unusual (Radchenko and Elmes 2001b). However, following a recent examination of the Myrmica of China (Radchenko and Elmes 2009a), we have shown that the pachei-group is much more diverse than previously expected. It now contains 14 species that are found in the Himalaya, south-western and southern China.

ritae-complex
Myrmica alperti

Myrmica angulata

Myrmica emeryi

Myrmica gigantea

Myrmica indica

Myrmica margaritae

Myrmica pararitae

Myrmica pulchella

Myrmica ritae

Myrmica serica

Myrmica sinensis

Myrmica titanica

Myrmica urbanii

Myrmica weberi

boltoni-complex
Myrmica boltoni

Myrmica collingwoodi

Myrmica martensi

draco-complex
Myrmica draco

Myrmica poldii

Myrmica schoedli

Myrmica yamanei

Diagnosis Workers: frontal carinae merging with the rugae that extend to the occipital margin, they do not curve outwards and do not merge with rugae that surround antennal sockets; frontal lobes slightly curved, frons wide, and frontal lobes not extended. Scape very long, often longer than head, smoothly curved at the base, without any trace of lobe or carina. Anterior clypeal margin slightly convex and distinctly notched medially. Posterior head margin has narrow collar-like ridge and posterio-ventral angles of head prominent. Petiole very long and low (PI1 > 1.35, often> 2.0); postpetiole fig-shaped (seen from above). Propodeal spines very long; propodeallobes sharply pointed apically. Males: scape of various lengths, but unfortunately males are known only for a few species.

Usually it is quite easy to recognise species from the ritae-group because their overall appearance, particularly their elongated petiole, fig-shaped postpetiole, very long propodeal spines and relatively long appendages are quite distinct from other Myrmica species. It is probably the strongest candidate among all the groups to warrant subgenus status (see above). Despite this, preliminary studies suggest that the behavior and ecology of its species is very similar to that of other species groups. Furthermore, it is the only morphologically-based group that clearly resembles extinct fossil species (see Chapter 3.4), although it should be remembered that the modern ritae-group species shared a common ancestor with species from the other Myrmica species groups 36 Ma, 4-8 million years after the known fossil species lived (see Jansen et al. 2010). At one time the ritae-group was considered "archaic" containing just a few "relict" taxa, but our recent work has shown it to be rich in species, currently we recognise 21 extant species from the temperate zones of high mountains in the Himalaya, Meghalaya, southern China, Taiwan, Vietnam, Burma, Thailand and even Indonesia. We are confident that considerably more ritae-group species will be described when the high mountains of this region are better investigated. We are able to separate the group into three species complexes, but these will probably be revised when the males of all the species are properly known:


 * ritae-complex

Diagnosis Workers: head, alitrunk and waist very coarsely rugose or/and reticulate, surface of head dorsum between rugae not punctated, smooth, appears shiny. Males: scape short, SI1 < 0.50. This complex contains the majority (14) of species in the group, also some of the largest Myrmica species found to date belong to this complex (i.e. M. gigantea and M. titanica), but not all species are large (for example M. weberi workers are about the same size as those of M. ruginodis):


 * boltoni-complex

Diagnosis Workers: head, alitrunk and waist finely sculptured, rugulose or even striated, surface of head dorsum between rugae distinctly punctated, appearing dull. Males are unknown.


 * draco-complex

Diagnosis Workers: head dorsum and alitrunk rugose, but waist finely striated and punctated; surface of head dorsum between rugae punctated. Males: scape of various lengths, SI1 < 0.50 or > 0.90. The sculpture of the workers of this species complex seems to be intermediate between the ritae- and boltoni-complexes, but the males are known for only two of the species (M. draco and M. schoedli), the former has long scape and the latter short one. The composition of this complex might be better understood when additional material, including males of the other species, is obtained.

rubra group
Myrmica arisana

Myrmica kotokui

Myrmica rubra

Myrmica ruginodis

Diagnosis Workers: frontal carinae curved outwards to merge with the rugae that surround antennal sockets, frontal lobes slightly curved, frons wide and frontal lobes not extended. Scape very smoothly curved at the base, not angled and with no trace of lobe or carina. Anterior clypeal margin convex, not prominent and without a medial notch.

Males: scape long, S12 > 1.00.

The group includes four species. Two of them (M. rubra and M. ruginodis) have a Transpalaearctic distribution, while the others occur in East Asia and Taiwan.

rugosa group
as per Radchenko and Elmes (2003)

(incomplete list of species)
 * Myrmica aimonissabaudiae
 * Myrmica ereptrix

M. ereptrix clearly belongs to the rugosa-group, as does its host M. aimonissabaudiae, despite the peculiar, gross development of its postpetiole (see Radchenko and Elmes 2001a).

scabrinodis group
as per Radchenko and Elmes (2003)

(incomplete list of species)
 * Myrmica bibikoffi
 * Myrmica hirsuta
 * Myrmica laurae
 * Myrmica scabrinodis
 * Myrmica vandeli

M. laurae is a social parasite that has the same combination of characteristics as the karavajevi-group except that its queens are significantly larger. However, we do not place it in this group because by the shape of its scape the queens clearly belong to the scabrinodis-group together with the other social parasites M. bibikoffi, M. hirsuta and M. vandeli. Recently, moderately strong evidence has been produced to show that M. vandeli may be a temporary social parasite of M. scabrinodis colonies (Elmes, Radchenko and Thomas, in press), in which case, all these social parasites belong to the same species-group as their host species.