Myrmecocystus species groups

AntWiki: The Ants --- Online

Snelling's series of taxonomic treatments of this interesting genus grouped species into subgenera and, in some cases, species groups within these subgenera. The work presented here is from the 1976 revision and a subsequent update of that work in 1982.

Snelling, R. R. 1976. A revision of the honey ants, genus Myrmecocystus (Hymenoptera: Formicidae). Nat. Hist. Mus. Los Angel. Cty. Sci. Bull. 24: 1-163.

Snelling, R. R. 1982. A revision of the honey ants, genus Myrmecocystus, first supplement (Hymenoptera: Formicidae). Bull. South. Calif. Acad. Sci. 81: 69-86.

There are a number of keys for the Myrmecocystus species. This includes a short key to the subgenera and three keys to species, grouped by subgenera.


Diagnosis. Worker and female with septemdentate mandibles; eyes slightly longer than second antennomere; bicolored, head and thorax reddish, gaster blackish (except one unicolorous orange species); wing membranes of sexuals without erect whitish hairs, but marginal fringe often present.

Description. Worker: Mandible with seven teeth; ocelli distinct; eye small, a little longer than second antennomere; erect hairs usually abundant, at least on vertex, thoracic dorsum and gaster, often on malar area, scape and legs as well; pubescence usually extensive, often sufficiently dense to obscure the integument on the first two or three terga; bicolored, head and thorax reddish (though often extensively infuscated) and gaster blackish (except in wheeleri which is uniformly orange-colored). Female. Similar to worker except for usual caste characters; IOD less than OOD; wings without erect whitish hairs on membrane; marginal fringe hairs often present. Male. Wings as in female; aedeagus with at least a few setae; mandible usually without defined denticles basad of apical tooth.

Type Species. Myrmecocystus melliger

Etymology. Endio (Gr., at midday) + dioktes (Gr., hunter), in allusion to the diurnal foraging activities of the included species.

The species assigned to this subgenus are all diurnal foragers, often highly predaceous on small arthropods. Except for one wholly orange-ferruginous species, all possess blackish gasters which contrast with ferruginous head and thorax in the worker. The head and thorax are often extensively infuscated in some species and may appear to be brown.

There are several species groups, but their definition is not easy. The most widely distributed group is the melliger group and it is possible to derive the remaining groups from it. One divergent line may have given rise to the romainei group. Another line may have given rise to the mimicus-flaviceps and mimicus-kennedyi series. Much additional study will be necessary before evolutionary patterns are clarified.

melliger group

Species belonging to the melliger group are large, densely pubescent, with numerous erect hairs on the malar area as well as elsewhere on the body. The erect hairs on the first three gastric terga arise from conspicuous poriform punctures (Fig. 23) which, in some lights, appear to be small conical papillae. Males of this group may possess one or two minute denticles basad of the apical mandibular tooth and a broad, longitudinal, median area of the first three terga is without erect hairs, or with the hairs much sparser than near the sides.

Although in Texas one species (placodops) occurs in non-montane situations, the species of this group are most often found at low and moderate elevations of desert mountain ranges. Nests are commonly located in stony soil and the entrance may be partially concealed by a stone. Only one nest of intonsus has been observed and that was situated in coarse-grained sandy soil at the edge of a large wash. Whether this is typical for the species is unknown. The placement of this species in the melliger group is provisional pending discovery of the male.

Two species, intonsus and semirufus, make conspicuous crateriform tumuli, but nests of the other species are most often without superstructures of any sort. The workers disperse the soil at random, well away from the entrance. Occasionally, low, semicircular mounds are built up.

The species of the melliger group are all active predators and scavengers. They have also been observed at flowers taking nectar. Repletes are known for all except intonsus.

mimicus group

These are medium-sized species in which the head and thorax of the worker are extensively brownish, the frons and occiput are smooth and shiny and the third tergum is without conspicuous appressed pubescence. The malar area has few or no erect hairs.

This group ranges from Trans-Pecos Texas to California and south to central Mexico. The two species appear to be largely scavengers and also gather large quantities of nectar and honeydew; repletes are known for both.

These are primarily lowland species, but follow alluvial fans onto the lower slopes of desert mountain ranges. Nests are often situated at the base of a shrub or amidst a clump of grasses. Tumuli thus are often somewhat irregular. When nests are located in open areas, the tumulus is most often a low, broad, regular crater.

kennedyi group

In these species the head and thorax are bright ferruginous; rarely is the posterior part of the thorax infuscated; the gaster is black, black and ferruginous or simply ferruginous. The frons is smooth and shiny, with no more than a few widely separated, obscure punctures. There are no more than five erect hairs on the malar area, usually fewer and the third tergum is conspicuously pubescent (except in kennedyi). Females of two species (kathjuli and wheeleri) have the penultimate segment of the maxillary palp much wider near the base than preapically; females of three species (kathjuli, nequazcatl and wheeleri) have the micropunctures of the first two terga irregularly spaced, often with the median area of the disc sparsely punctate.

This is a wholly western group and ranges from southern Oregon and Idaho to central Sonora and northern Baja California. Nests are populous and the species are aggressive predators on small arthropods. They also gather large quantities of nectar and honeydew; repletes are known for two species. The nests are often found in fine, but well compacted sand, in clear areas. A large, crateriform tumulus is present.

flaviceps group

Closely related to species of the foregoing group and perhaps should be included there. Workers differ, however, in being duller. The head and thorax may be dull ferruginous, but often extensively infuscated. The frons is usually closely, finely punctate. The female is extensively infuscated on the head and thorax and has the first four terga shiny, with scattered coarse, setigerous punctures on the discs.

This ant ranges from southwestern Utah, southern Nevada and adjacent California south to Sonora and central Baja California Sur. It is largely a scavenger and regularly visits flowers for nectar. Repletes are known. Nests are found in a wide variety of situations, from loose sand to dense, clayey soil. A low, circular crateriform tumulus is present.

romainei group

This group is close to both the kennedyi and flaviceps groups, but the workers possess six or more erect hairs on the malar area. Erect body hairs are more numerous than in species of the other two groups. The frons has conspicuous rather coarse punctures, except in the smallest workers. The two basal terga of the male are without pubescence on the discs, but with sparse pubescence laterad.

The group ranges from western Kansas and Oklahoma to southern Nevada and adjacent California. The more widely distributed species, romainei, is found in a wide variety of habitats. This species may locate nests in fine, deep sand or in compact clayey soil. Nests of koso are usually in compact, rocky soils. Both species construct large, crateriform tumuli.

Little is known of the habits of these species. They are predators and scavengers. Probably, too, they visit flowers for nectar. Repletes are not known.


Diagnosis: Worker and female with septemdentate mandibles; eyes little, if any, longer than first flagellomere; wings of sexuals with numerous fine, erect white hairs on membrane; small concolorous blackish or dark brownish species.

Description: Worker: Mandible with seven teeth; ocelli distinct; eye small, maximum length hardly exceeding that of first flagellomere; clypeus about twice wider than long; fewer than six hairs on malar area in frontal view; pubescence reduced, not obscuring tergal surfaces (exception: colei, n. sp.) and very sparse on head; monochromatic blackish to brownish, lower third of head often yellowish. Female: Similar to worker, except for usual caste characters; wings with abundant fine, erect whitish hairs on membrane; wing fringe absent (exception: creightoni Snelling). Male: Wing membrane with abundant fine, erect, whitish hairs; aedeagus without setae; mandible usually without defined denticles basad of apical tooth.

Type Species: Myrmecocystus creightoni

Etymology: Eremnos (Gr., swarthy, dark) + Kystis (Gr., bladder), reflecting the uniformly dark color of these honey ants, their most important field recognition characteristic.

This subgenus includes all the small, uniformly dark species. Their small size (largest workers 6 mm or less in length) and distinctive coloration render them easily recognizable in the field. They superficially resemble the dolichoderine species, Conomyrma insane (Buckley).

Eremnocystus is the least widely distributed subgenus of Myrmecocystus. One species is found in southwestern Idaho and I have samples of what may prove to be another species from Washoe County, Nevada. The bulk of the species occur in southern California. Two species are found in the Mojave Desert, two are found in the Colorado Desert, one occurs in the valleys between the Transverse Ranges and one is known from the coastal sand dunes around Bahia San Quintin in Baja California.

The treatment afforded the group here is not wholly satisfactory. The small size, obscure habits and disjunct ranges of the component species have resulted in their rarity in collections, even though the species may be quite common where they occur. Most of the species are wholly allopatric with one another. The only area of known sympatry is that of Yuma, Arizona, where tenuinodis and yuma are found, often nesting within a few meters of one another. Adjacent, but apparently wholly aIlopatric species are lugubris, creightoni and colei. Both hammettensis and perimeces appear to be well outside the known ranges of other species.

The current distribution pattern suggests the possibility that these populations could be treated, for the most part, as subspecies. However, the lack of material of an intermediate nature and the constancy of characters within a population do not support such an interpretation. Too, characters used here to differentiate the species are of the same magnitude as those used elsewhere in this revision.

It appears that the subgenus may once have had a wider, and more continuous, distribution and probably consisted of a single species, such as creightoni. With increasing aridity in the southwestern United States populations may have become isolated and evolved along restricted lines which became increasingly exaggerated in the absence of disruptive gene flow from other populations. No doubt, too, some selective pressure has been exerted by the larger, and more aggressive, diurnal foragers of the subgenus Endiodioctes. The largest and hairiest member of Eremnocystus is colei. This species is found in the Los Angeles Basin and the Cajon Canyon area of San Bernardino County. In this area the only competing diurnal forager is wheeleri in the subgenus Endiodioctes. In southwestern Idaho hammettensis shares diurnal foraging with kennedyi, another species of Endiodioctes. Slightly larger size and abundant erect pilosity are characteristic of hammettensis.

Few data are available on the foraging habits of hammettensis. The one colony which I found in Owyhee County, Idaho, was taken before noon on a warm day (ambient temperature approx. 85°F.). Workers were foraging on nearby plants at this time and active excavation within the colony was indicated. The habits of colei are somewhat better known. This species is an active forager and may be found foraging during the warm midday hours at ambient temperatures up to about 90°F. Both of these species, then, are active out of the nest at midday.

Other species of Eremnocystus forage in the cooler morning and afternoon hours. They usually emerge shortly after sunrise, dispersing rapidly over the surface and working into low vegetation. At least one species (tenuinodis) may develop short foraging files to a concentrated food resource. Surface activities continue until the ambient temperature reaches about 80°F and by the time the temperature rises to 85° all foragers have returned to the nest. Soil surface temperatures of about 100°F elicit a pronounced stress reaction and temperatures in excess of 110° are quickly fatal.

During the hot middle portions of the day, the ants remain away from the surface, usually retreating to depths of 10-15 cm. Often the entrance is blocked by soil particles. When temperature has dropped to a tolerable level in late afternoon or early evening, foraging activity is resumed until full darkness, by which time all workers have again returned to the nest. In the southern deserts the hottest part of the year is usually from the end of June to about the middle of September. During this season conditions may not be suitable for days or even weeks, and the ants may wholly curtail surface activities. Aestivation during the hottest summer period appears to be normal for these species.

Those species of Eremnocystus which appear to be wholly matinal and crepuscular are all characterized by a reduction in the erect body hairs and in appressed pubescence. These species also appear to be relatively short-legged, though I have not made measurements to confirm this impression. Relatively long legs and an abundance of erect and appressed body hairs seem to be adaptations which are favorable to ants active during the hot periods of the day. The hairs and pubescence are highly reflective and presumably function to deflect both heat and light. The diurnal foraging species, when moving across a very hot surface (one in excess of 110°F) dart rapidly from one raised object, pebble, twig, etc. to another. Upon reaching such an object, the ant stops atop it for several seconds. During this process the body is held as high from the substrate as possible, often with the gaster elevated. Such actions take advantage of the sharp reduction in temperature within a few millimeters of the surface. The two diurnal species of Eremnocystus, colei and hammettensis, do possess an abundance of erect hairs and appressed pubescence. Their legs appear to be relatively longer than in the other species of Eremnocystus and they exhibit heat avoidance reactions as discussed above. None of these characteristics apply to the remaining Eremnocystus.

Eremnocystus appears to be a derivative of Endiodioctes, through reduction in body size, pubescence and pilosity and loss of the wide range of polymorphism evident in that subgenus. These smaller species are less aggressive foragers and their occupation of the matinal-crepuscular temporal niche may be a response to competitive pressure by the species of Endiodioctes. It is possibly significant that the matinal-crepuscular species occur in areas where there are two or more species of Endiodioctes. The two Eremnocystus species which forage diurnally both are found in habitats where there is but a single species of Endiodioctes.

No attempt has been made here to divide the species of Eremnocystus into species groups. Two species, colei and perimeces, might form monotypic groups. Another group might consist of lugubris and tenuinodis and a fourth group would include the remaining species, ereightoni, hammettensis and yuma. However, I do not feel that I know the species sufficiently well to be satisfied that these groups would be natural ones. When the sexual forms are better known it may be possible to define species groups.


Key to species

Diagnosis. Worker and female: With mandible basically nine-toothed; eye large, EL usually conspicuously greater than length of first flagellomere; wings of sexuals without fine erect hairs on membrane; male mandible often denticulate; worker and female concolorous light yellow to brownish yellow, male light to medium brown.

Description. Worker: Mandible basically nine-toothed, but may be eight- or ten-toothed; eye large, always exceeding length of first flagellomere; ocelli much reduced or absent; pubescence and erect hairs variable, sparse or abundant; integument basically yellowish, may be extensively infuscated so ant appears light brown, but never ferruginous and/or blackish. Female: Similar to worker except for usual caste characters; ocelli present, distinct; wings without erect white hairs on membrane. Male: Wings as in female; mandible normally with at least one denticle basad of apical tooth, often two or three; aedeagus without setae.

Type Species: Myrmecocystus mexicanus

Discussion. To this subgenus are assigned the distinctly yellowish nocturnal species. In two species (melanoticus and mexicanus) the yellow color may be extensively overlaid with brown, but even in the darkest of these there is no approach to the condition of the species of Endiodioctes or Eremnocystus.

The basic number of mandibular teeth is nine. Such species as melanoticus, mexicanus and navajo' normally possess nine-toothed mandibles, but there is variation in these from eight to ten. The usual number of teeth in testaceus is eight and in the two derived species, ewarti and pyramicus, it is seven. Variation in the three latter species usually is expressed by the presence of one or two additional denticles (one in testaceus, two in ewarti and pyramicus).

Eye length is also somewhat variable, but it always exceeds that of the first flagellomere. In a few species (mexicanus, navajo, testaceus) it may be as little as 1.05 times the length of the first flagellomere, but, even in these species, is normally in excess of 1.2 times. The eye is relatively larger in the two species ewarti and pyramicus, which I consider to be most derived, at least 1.5 x the first flagellomere. In these same two species the OMD is shorter than in species of the other groups, not exceeding I. 15 x EL.

mexicanus group

Within this subgenus the most widely distributed species is mexicanus and this is apparently the most primitive. The mandibles of the worker possess nine teeth and the most common variant form of dentition exhibits an increase to a decemdentate condition. Far less commonly, an octodentate variant appears in some colonies. Very closely related to mexicanus is the southern species melanoticus. In this species the basic number of mandibular teeth is still nine, but variants with eight teeth are common, while a ten-toothed variant is very rare. This species is as hairy as mexicanus and similar to it in size, but apparently a little less polymorphic. A third species closely related to these is navajo. It is smaller in stature and a little less polymorphic than mexicanus. In navajo the vestiture is reduced, especially the erect hairs of the appendages. The mandibles are most commonly with nine teeth, but some individuals exhibit a reduction to eight in one or both mandibles. The most striking deviation from mexicanus lies in the displacement of the eyes toward the top of the head; in most workers the upper eye margin is coincident with the occipital margin in full face view.

pyramicus group

The pyramicus group consists of the two species ewarti and pyramicus. The mandibles of these two species are basically septendentate, with one or two smaller denticles sometimes present, the pilosity is greatly reduced, especially on the thorax, and the eyes are displaced toward the top of the head. The eyes are large, consistently at least 1.5 x the first flagellomere. In profile, the metanotum is greatly depressed; the basal face of the propodeum is flat and slopes to the metanotum, so that the juncture of the basal and posterior faces appears to be projected upward. These are the smallest species in the subgenus and are limited to the western range of the subgenus.

testaceus group

The third species group contains but a single western species, testaceus, which ranges from southern Washington to northern Baja California. This ant is fully as hairy as mexicanus, but is a little smaller, more robust and less polymorphic. The mandibles are basically octodentate, but one additional tooth is sometimes present. The eyes are large, usually about 1.25 x the first flageliomere, but ranging from 1.05-1.40. In this species the metanotum is not depressed (depressed in mexicanus group species) and the propodeum is higher than long, with the juncture of the basal and posterior faces abruptly rounded, sometimes subangulate.

Unplaced to group

Appears to be annectant between the mexicanus group and the pyramicus group.

The evolutionary sequence of this subgenus may have started with a widely distributed mexicanus-like form. A tendency toward loss of a single mandibular tooth and some reduction of polymorphism would result in melanoticus. The navajo form is also a straightforward derivative through reduction in size and in amount of vestiture; the eyes are displaced toward the top of the head. Progression of the same features seen in navajo results in the ewarti and pyramicus characteristics in which eye size and reduction of vestiture are most ex treme. The characteristic thoracic profile of these two species is an exaggeration of characteristics present in mexicanus and foreshadowed in navajo. An independent line appears to be that of the monotypic testaceus group. Here, the number of mandibular teeth is reduced to eight, but the body is as hairy as in mexicanus, the eyes are about as large as in mexicanus and are not displaced dorsad. However, the thorax is stouter, the metanotum is not depressed and the propodeum is higher than long, with the dorsal face flattened.

Predictably, those species which are widely distributed are also those which occur in the widest range from Pinon-Juniper Woodland to Saltbush-Greasewood Shrub. On the other hand ewarti is a species of Creosote bush Shrub and Creosote bush-Bur sage Shrub and pyramicus is found largely in Sagebrush Steppe.

All of the species forage mostly between sundown and sunrise. They may emerge on overcast days, as testaceus commonly does. Foraging is usually initiated as a mass exodus in which workers quickly scatter in all directions. Aphids and pseudococcids are solicited for honeydew and the exudates from cynipid galls are gathered. These are supplemented by visits to floral and extrafloral nectaries. Protein sources are largely the scavenged remains of other arthropods; the ants rarely function as predators. When occasion permits they will scavenge on dead vertebrates. All species are known to form repletes.

Refuse chambers are rare in nests. Debris from the nests is usually carried to the surface and dropped at random some distance from the entrance. Often it is dropped from the summit of the crateriform superstructure.

Tumuli are normally present, as regular craters, and consist of the coarsest grains available. Fine sand is usually scattered at random beyond the tumulus. Tumuli of melanoticus, mexicanus and navajo examined immediately after rains have been noted to be little worn and the particles may be agglutinated.