Melophorus

These ants are common in most areas of Australia with the exception of cool, wet areas. Foraging is strictly during the day, and in the arid zone they are one of the only ants active during the hottest part of the day. Some species reduce foraging activity during the winter, although in cooler areas most species will forage on warmer winter days. They are omnivores and some species will collect seeds (Briese and Macauley 1981). Some species have become highly modified with very flattened bodies, presumably to allow foraging in small, thin crevices (Greenslade 1979). A few species are also known to prey on the larvae and pupae of meat ants in the genus Iridomyrmex (Clark 1941). All species are ground nesting in the open, often with a small amount of dirt near the entrance. Workers of Melophorus are some of the fastest and most timid ants in Australia. On hot days they can run so rapidly that they can be difficult to capture. They are also so shy that the slightest disturbance near a nest will cause workers to disperse or disappear into the nest, only returning or becoming active again after many minutes.

Identification
The spiracle on the propodeum is elongate and slit-like. The underside of the head and mandibles usually have numerous elongate hairs although in some species these hairs are reduced to only 2 or 3 on the under side of the head and are absent from the mandibles. A small opening which is usually fringed with hairs (the metapleural gland) is present above the hind leg.

Melophorus is one of the most abundant and diverse groups of ants in the arid zone of Australia. They are most similar to some species of Camponotus in general body size and shape, however they differ in having a small opening above the hind leg and in having the propodeal spiracle much longer and narrower.

Heterick et al. (2017) - Worker Polymorphic ants characterized by a combination of long, J-shaped setae on the mentum, an elongate (often slit- or comma-shaped) propodeal spiracle, presence of a metapleural gland, antennal insertions abutting the posterior clypeal margin and three ocelli in an ocular triangle in all workers. The genus is taxonomically compact, and with the above characters taken into consideration, its members cannot be mistaken for any other group of ants. The descriptions below relate to the small minor worker and the large major worker. Media workers are of intermediate appearance but more closely resemble the minor worker.

Queen Larger to smaller than the conspecific major worker. Palp formula, ocelli, appearance of antenna and number of antennal segments, and number of mandibular teeth are as for the conspecific major worker. Mesoscutum about as wide as long in species seen. Parapsidal lines parallel and usually distinct. Axillae small and widely separated, the separation much greater than their width. The wing is of the ‘Camponotus’ form, with a radial cell, a first cubital cell and one submarginal cell. The cross-vein cu-a is present but the discoidal cell and cross vein m-cu are absent. Ergatoid or brachypterous queens have not been identified.

Male (Caution: this diagnosis is based on only a small number of specimens that we have been able to identify as Melophorus males). Males with typical characteristics of males of this subfamily. The head bears an ocellar triangle that may be weakly turreted in some species, the antennal scape is usually longer to much longer than the head and the 13-segmented antenna is elbowed. The eyes are generally larger than those in the corresponding queen or major worker but not globose or bulbous in males seen. The mandible is 1- to 4-toothed in males seen; if 1-toothed, then apical tooth is prominent. The wing is as for the queen, but in one species a second small submarginal cell may be present, possibly limited to certain individuals. The parapsidal line is always present, weakly to strongly impressed; the notaulus is absent or present only as a median appendix arising from the base of the mesoscutum. The axillae are small and very widely separated from one another, the separation much greater than their width. Worker-like, wingless males are known for at least one desert-dwelling species.

The species are arranged into species groups: Melophorus species groups

Distribution
These ants are only known to occur in Australia. They are found in all areas except the higher elevations of the south-east and Tasmania, although they are by far the most abundant and diverse in the arid and semi-arid zone. They can be found in all of the drier habitats, being much less common or absent from wet sclerophyll, especially in southern regions, and they are absent from rainforests.

Heterick et al. (2017) - The genus Melophorus is entirely restricted to Australia, and all known members are thermophilic or believed to be predominantly so, mainly being terrestrial foragers. The very many species of Melophorus are particularly abundant in hot, arid or semi-arid environments where they are diurnally active, mostly in the summer months.

Biology
Heterick et al. (2017) - Although the genus Melophorus is generally poorly known, several species, most notably Melophorus bagoti, have been the focus of concentrated research. Much of the current interest in M. bagoti concerns homing strategies in this ant (e.g., Wehner et al. 2006, Narendra 2007a, Narendra 2007b, Legge et al. 2010); while more occasional research has covered nest structure and activity (Conway 1992), thermophilism (Christian and Morton 1992) and general biology and taxonomy (Wheeler 1908). Other Melophorus species that have received individual attention have been Melophorus perthensis (as ‘Melophorus turneri perthensis’) (Majer et al. 2010), an unidentified species in the Melophorus aeneovirens group (Hoffman 1998) and those taxa that feature in taxonomic papers with some commentary on the habits of the species described (e.g., Melophorus fulvihirtus, in Clark 1941, and Melophorus anderseni and Melophorus majeri, in Agosti 1997).

In terms of general role in the environment, Melophorus ants are a highly thermophilic group that fills the same ecological niche as Cataglyphis in the Old world and Myrmecocystus in North America (Andersen 2007). All species are active only during the day, mostly at peak diurnal temperatures and predominantly in the summer months. Briese and Macauley (1980) found that one species (‘Melophorus sp. A’) studied at Emmet Vale, NSW, was only active when the temperature was between 37°C and 54°C, and this species was inactive for prolonged periods in the cooler months. Species are richest in arid or semi-arid environments, and relatively few inhabit thick forest or cooler, mesic areas. There are no alpine specialists. Typically, Melophorus species nest directly in soil and can rarely be found under stones or woody debris. Although they may forage on tree-trunks and vegetation no arboreal nesters are known. Soil nests of very common species such as Melophorus turneri and Melophorus perthensis are easily recognizable by the small, crescentic dunes of excavated sand that surround the nest hole (senior author pers. obs.). Melophorus nests are often or always closed with sand or pebbles at night (McAreavey 1947, current authors pers. obs.). Members of this genus are exceptionally timid ants, avoiding conflict with other ants (e.g., meat ants) by occupying a different temporal niche. They will also retreat into their nest upon the slightest human disturbance, and will not reemerge for several minutes (Shattuck 1999).

Melophorus ants have a diverse range of diets, depending on the species, and some are quite specialized. Melophorus potteri is a termite raider (McAreavey 1947, Andersen 2000), and its bluntly serrate mandibles with a single large apical tooth may represent an adaptation to this lifeway. Interestingly, raids are carried out by individual workers, and not columns of workers, as in many formicid termitophages. A second species in the M. potteri species-group (Melophorus pelecygnathus) has extraordinary hatchet-shaped, edentate mandibles and may have similar behavior. However, the third species in this group (Melophorus macroschismus has the normal 5-toothed dentition of most Melophorus. Melophorus anderseni and Melophorus fulvihirtus raid meat ants’ (Iridomyrmex purpureus species-group) nests. Clark (1941) mentions the raiding activity of M. fulvihirtus in a brief note, but Agosti’s (1997) observations on M. anderseni are outlined in detail: this latter species uses subtle behavioural characters and the scent of its host so that it can carry off meat ant pupae. Workers of this ant were seen by Agosti to actually rub themselves against meat ant workers in order to obtain their nest odour.

The M. wheeleri species-complex, in which the major workers have powerful, truncate, inwardly curved mandibles, are said to be obligate granivores (Andersen 2007). In some cases, exotic plants with suitable seeds may have replaced the use of native species in the diet of members of this species-group: for instance, in southern WA, Melophorus wheeleri is known to feed on the seeds of the exotic grass Avena barbata (Harris and Standish 2008).

Apart from such specialized forms, most Melophorus are omnivores but include a large amount of vegetable material in their diet. Berg (1975) concluded that Melophorus species have a strong preference for elaiosomes, the nutrient rich appendages found on the seeds of many Australian plants. Elaiosome-eating species such as Melophorus perthensis are likely to play an important role in native seed germination (Majer et al. 2011). Melophorus perthensis is one of the major seed harvesters in the West Australian jarrah forest (Majer 1982). Davidson and Morton (1981), who examined seed removal in inland NSW, found that diaspores (seeds modified for dispersal) moved by ‘Melophorus sp. H’ tended to be in the smaller range (≫ 4.09 mg). Animal protein in the case of omnivores may be supplied in the form of insects roasted in desert heat (Narendra 2007a) or dead ants discarded by their nestmates (Briese and Macauley 1981). Such scavenged Formicidae can constitute a surprisingly high percentage of food items: Briese and Macauley’s study includes tabular information that 40% of the food items taken back to their nest by an unnamed Melophorus species (‘Melophorus sp. A’) were of this nature. Several gracile species forage on the trunks of eucalypts. Greenslade (1979) states that one such species, with a peculiarly flattened head, is adapted to foraging under bark. Melophorus may attend lycaenid butterfly larvae that produce sugary glandular secretions, but known associations of lycaenid larvae with this genus of ants appear to be relatively minor and purely facultative (Eastwood and Fraser 1999, Fiedler 2001).

In turn, Melophorus species, especially their alates, are an important food source for some agamid lizards (e.g., they are one of the few sources of food for the Lake Eyre dragon lizard Ctenophorus maculosus Mitchell [Chan et al. 2009]), other ants (e.g., Iridomyrmex) and spiders (a zodariid spider is mentioned as an important predator of Melophorus bagoti by Muser et al. 2005). On the other hand, non-agamid lizards prefer termites over ants (Abensperg-Traun and Steven 1997). Information on Melophorus as a dietary item for other Australian vertebrates is sketchy at best and usually lacking. Melophorus species are rare in the diet of numbats (Calaby 1960), and there is little or no specific information on genera of ants, other than Iridomyrmex, included in the food intake of other ant-eating mammals such as echidnas and several smaller carnivorous marsupials that will eat ants. Insectivorous snakes may take Melophorus, although there is just the one record of Melophorus as a prey item (for Ramatyphlops australis) in the list of ant genera eaten by blind snakes (Typhlopidae) supplied by Webb and Shine (1993). Melophorus alates, presumably, also run the risk of predation by birds. Worker ants, however, are probably not taken by most birds because of the uncomfortably high temperatures at which they are active and their speed over the ground surface. Replete workers of Melophorus bagoti are famous as food for desert dwelling Aboriginal people, but some references to their use are qualified (e.g., Conway 1992) or reject that repletes of M. bagoti are acceptable as human fare (thus, Australian Myrmecology Gallery 2002).

Homing strategies. Melophorus bagoti (and, in all likelihood, other desert-dwelling species of Melophorus) uses a type of homing device called path-integration that enables ants to return to their nest along the shortest return route in a featureless environment (Narendra 2007a). Presumably, this type of orienting also compensates for the difficulty in following pheromone trials laid on shifting or unstable, windblown soils. Ants may also be blown off their paths by dust storms (Narendra 2007b). In a carefully manipulated experiment, Legge et al. (2010) pursued this question further. Where path integration information was not available, M. bagoti workers did not use visual landmarks (artificially provided in the study). Instead, they oriented in a particular direction (in this case the exit from an artificial arena) regardless of whether or not it was aligned with the ants’ nest. Thus, it appears that vectors used by the ant to return to its nest are based on compass information gleaned from previous trips, i.e., the path integrator does not process information based on the current journey. The nature of the landmarks that enable these ants to learn local vectors in natural circumstances have not been determined as yet, but Legge et al. surmise that possibly the skyline is a factor, as is polarized light.

Biomonitoring Potential. In Australia, Melophorus ants are currently an important component in the Ant Functional Group (e.g., Andersen 1990, 1995) mix used by researchers and land managers to investigate the success of minesite, rangeland and burnt area rehabilitation. However, their thermophilism also makes them a potential candidate in investigations into microhabitat alteration as climate and weather patterns change under the influence of global warming. Hypothetically, the natural range of many species may be expected to be extended with increasing aridity and higher average temperatures, particularly in the south west of the Australian continent.

As a final note, in the field most members of this genus of ants are extremely difficult to catch because of their rapid movement across the ground. The senior author found that major workers and minor workers of large species could be slowed without damage by dabbing gently at them with cotton wool impregnated with ethanol, and then picking the ants off the cotton wool with forceps. Minor workers were collected by suctioning – mainly at the entrance to their nest – with a pooter and transferring them immediately to an ice bath (i.e., a basin filled with ice and water). This had the effect of paralyzing these thermophilic organisms so that they lay quiescent on the water surface and could be transferred with ease to a vial of ethanol.

Nomenclature

 *  MELOPHORUS [Formicinae: Melophorini]
 * Melophorus Lubbock, 1883: 51. Type-species: Melophorus bagoti, by monotypy.
 * Melophorus senior synonym of Erimelophorus, Trichomelophorus: Brown, 1955d: 474.
 * ERIMELOPHORUS [junior synonym of Melophorus]
 * Erimelophorus Wheeler, W.M. 1935c: 71 [as subgenus of Melophorus]. Type-species: Melophorus wheeleri, by original designation.
 * Erimelophorus junior synonym of Melophorus: Brown, 1955d: 474.
 * TRICHOMELOPHORUS [junior synonym of Melophorus]
 * Trichomelophorus Wheeler, W.M. 1935c: 71 [as subgenus of Melophorus]. Type-species: Melophorus hirsutus, by original designation.
 * Trichomelophorus junior synonym of Melophorus: Brown, 1955d: 474.

Heterick et al. (2017) - Taxonomic history. The genus Melophorus was originally erected for the single species Melophorus bagoti. However, a species described earlier in the genus Formica was later revealed to be a Melophorus (Melophorus aeneovirens). Lubbock’s diagnosis of the genus was poor and initially Melophorus lacked any real systematic framework. As a result, Melophorus was initially very broadly defined and accumulated species, including non-Australian species, that are now placed in other genera.

Forel (1911) had already commented on the diverse morphology of Melophorus that caused many of the then known species to be confused with unrelated genera or extralimital genera (i.e. those occurring outside of Australia). However, based on the appearance of the proventriculus, he affirmed their basic unity at the generic level. In 1912, he made a more determined effort to formally link the overly broad ‘Melophorus’ with taxa that seemed to him to share significant apomorphies (Forel 1912). In this work, ‘Melophorus’ was placed in a subfamily titled Camponotinae (originally erected by Forel in 1878 as ‘Camponotidae’ and later [1893] corrected by him). Forel divided the Camponotinae into three assemblages of taxa, based principally on the proventriculus. He also provided all-caste keys to the formicine genera known at the time. ‘Melophorus’ found a home in the second assemblage, the so-called Mesocamponotinae, and in a tribe, Melophorini, that consisted of Melophorus and another Australian genus, Notoncus. These two genera were distinguished from other formicines by their 12-segmented antennae and (with the exception of the ‘Australian Melophorus’!) the presence of a discoidal cell in the wing of the sexuals.

Within the first two decades of the twentieth century most of the species wrongly placed in Melophorus were correctly reassigned by Forel and another great myrmecologist of his era, Carlo Emery. Similarly, species such as M. aeneovirens were incorporated into Melophorus. Wheeler (1920) replaced the old subfamily name Camponotinae with Formicinae, based on his work on larval characters, and established the foundations of a subfamily classification that remained substantially unchanged (with just a few additions) until the modern era of molecular systematics.

By 1922, the taxonomic status of Melophorus as a well-defined genus in its own right was gradually starting to emerge, although in that year Carlo Emery still subsumed genuine Melophorus species (along with several Prolasius species and Notoncus spinisquamis Andre) under a subgenus ‘Melophorus’ (Emery 1922). ‘Prolasius’ and ‘Lasiophanes’ constituted the remaining two subgenera under the broader umbrella genus Melophorus. Interestingly, Notoncus was treated by Emery as a discrete, though closely related genus. In Emery’s treatment, all non-Australian taxa had now been removed from the subgenus ‘Melophorus’, although of the diagnostic characters that Emery listed for workers of the subgenus, only polymorphism would now be recognized as a useful character.

William Morton Wheeler (1935) made another major advance in rearranging the structure of the Melophorini, an arrangment largely retained in modern thinking on the group. Lasiophanes, Melophorus and Prolasius (the last including, however, two Notoncus species) assumed full generic status. The remaining genera in the tribe Melophorini were Diodontolepis (including Notoncus spinisquamis only), most of the remaining Notoncus under the genus of that name, Pseudonotoncus and Myrmecorhynchus. Wheeler removed a number of names from the existing genus Melophorus and reassigned them to Prolasius. Wheeler also tackled variation within Melophorus by erecting three subgenera: Melophorus sensu stricto for the gracile forms like M. aeneovirens that resemble the American honey-pot ants in the genus Myrmecocystus, Erimelophorus for the large-headed forms and Trichomelophorus for the peculiar Melophorus hirsutus.

The last major contribution to the taxonomy of Melophorus in its own right was made by Brown (1955). He dismissed Wheeler’s three subgenera on the basis of structure and habits and formally synonymized them. He also drew attention to two important diagnostic characters for the genus, namely, the long ‘ammochaetae’ (psammophore) together with the J-shaped setae on the mentum and also the elongate propodeal spiracle.

Since 1955 there has been no serious taxonomic consideration of the genus. Andersen (2000), in a field guide, provided a key to northern Melophorus that included very many groups. These groups appear to have been chosen on phenetic grounds, and the characters used, e.g., sculpture, pilosity, etc. were mainly those that have little validity at the species-group or subgenus level. A few characters used, such as length of the palps, have more significance, and they are discussed later in this work. A later discussion (Andersen 2007), evidently developed from the first (i.e., Andersen 2000), and covering a broader suite of Melophorus, is much more substantial, and (to our minds) accurately identifies some apomorphies that set apart several, although not all, significant species-groups that we recognize. In this case, Andersen subsumed such groupings under broader ‘radiations’. His identification of the Melophoprus fulvihirtus, the Melophoprus potteri and the Melophoprus aeneovirens species-group are supported in this work, as, in general terms, are the characters used to differentiate them, viz., appearance of the palps, the presence or absence of metatibial apical spurs, and the clypeal apron (in the aeneovirens species-group). Nonetheless, not all the characters mentioned are correctly applied (e.g., the metatibial spur is not absent in M. fulvihirtus, as stated by Andersen). Other important characters, such as the placement of the clypeal psammophore, the shape and dentition of the mandible and the nature of the preapical metatibial spines, are discussed in this work.

Apart from Andersen’s two works, there has been little else on the genus since 1955. Two rather aberrant and interesting Melophorus, Melophorus anderseni and Melophorus majeri, were described by Agosti in 1997, the first new species to be described since Melophorus bruneus in 1949. No Melophorus species have been described since 1997.

At a higher taxonomic level, Agosti (1991) used the unfused helcium (the collar-like pre-tergite and pre-sternite of the third gastral segment) and the close alignment of the hind coxae to place Melophorus in the Formica species-group. However, Bolton (2003) reverted to the traditional morphological argument based on the anatomy of the proventriculus, along with details of the mandible and metatibial setae, to situate Melophorus as the sole genus in the Tribe Melophorini. Recent molecular work (Moreau et al. 2006 using five nuclear genes and one mitochondrial gene) strongly supported an Australoid formicine clade comprising Myrmecorhynchus, Melophorus, Prolasius and Notoncus, confirming the suspicions of the early Twentieth century myrmecologists such as Forel, Emery and Wheeler. Finally, Ward et al. (2016), in a molecular phylogenetic study, posited a tribe Melophorini that consists of eight purely Australian genera (Melophorus, Myrmecorhynchus, Notoncus, Notostigma, Prolasius, Pseudonotoncus, Stigmacros and Teratomyrmex) and one South American genus, Lasiophanes. Interestingly, this paper confirms much of Wheeler’s earlier (1935) analysis, especially when one considers that the uncommon and localised genus Teratomyrmex had not yet been discovered.

The current position of Melophorus. This genus is currently very poorly characterized considering its importance in the Australian landscape. At the time of writing there are only 24 full species and eight subspecies in the taxonomic literature. One taxon (Melophorus cowlei Froggatt, here reinstated as a junior synonym of Melophorus bagoti) was misidentified for quite a large portion of its history as a Camponotus. The lack of taxonomic activity, especially in regards to the smaller species, is quite puzzling. Synonymy is extensive, and clusters around a small handful of the larger and more ubiquitous taxa (namely, Melophorus aeneovirens, Melophorus biroi, Melophorus turneri and Melophorus wheeleri). Much of the descriptive work took place in the era 1900-1930, which corresponds to the most productive period of the great continental taxonomist-systematist Auguste Forel, who described almost fifty per cent of the Melophorus fauna prior to this monograph. After 1930 the description of new Melophorus species has largely been the province of the amateur researcher Father John J. McAreavey and the economic entomologist John Clark.

Description
Heterick et al. (2017):

Worker
Minor Head. Head oval, square, cordate or rectangular, its posterior margin strongly convex to strongly concave. Compound eyes always present, usually medium to large (eye length generally 0.2× length of side of head capsule ≥), rarely small. Eye shape elliptical across most species, often slightly reniform due to an invagination of cuticle on the outer margin of the eye, much more rarely ovoid, roughly circular or oval tending to elongate; in some populations of M. hirsutus eyes flattened and embedded in cuticular prominences when seen in full-face view, or projecting roundly or even bizarrely conical. In full-face view, eyes mostly situated in the upper half of the head capsule, rarely (mostly in very small minor workers) at about the midpoint of the capsule, but never anteriad of the midpoint. In profile, eyes usually placed ahead of the midline of the head capsule, less frequently at about the midline or fractionally in front of it, but never behind the midline. Three small ocelli always present. In full-face view, frontal carinae convex, concave, straight, convergent posteriad or divergent posteriad, sometimes raised (particularly near antennal insertion), vestigial beyond antennal insertions in a few taxa. Torulus (antennal sclerite) pedunculated in a couple of species but otherwise inconspicuous. Antenna always 12-segmented. Scape very variable in length, but usually attaining the posterior margin of the head capsule, and often extending well beyond it. Clypeus never longitudinally carinate, or with protruding clypeal teeth but otherwise variable in appearance. Posterior margin of clypeus mostly arched and falling away from its mid-sector (i.e., between the frontal carinae), but lateral sectors of clypeus on same plane as mid-sector in a few species. Anterior clypeal margin often convex, with or without an anteromedial dimple, but concave, acuminate, spatulate and variously protrusive clypeal morphologies also occur. The midpoint of the clypeus may be notched, especially in the aeneovirens species-group. A clypeal psammophore of long setae (or ammochaetae) stretched across that sclerite is found in all Melophorus species, and may be located at any elevation, from at or just above the anterior clypeal margin to just under the posterior clypeal margin. Mandibular and gular psammophores of long, curved setae also usually present. Mouthparts. Palp formula mostly 6,4, but 3,4 and 3,3 also occur, maxillary palps variable in length from not reaching the hypostomal bridge to attaining the mesopleuron when they are fully extended. Apical segment of maxillary palp usually oblongiform, but may be narrow and acuminate or even vestigial and barely visible under a stereomicroscope. Mandible most commonly with five distinct teeth, but number of teeth and denticles ranges from four to over 15, mandible edentate except for apical tooth in one species. In five-toothed species apical tooth is nearly always much longer than remaining teeth and teeth 2 and 4 are typically longer than tooth 3 (teeth count taken from apical to basal tooth), basal tooth often offset and directed vertically. Masticatory margin usually vertical or weakly oblique, but strongly oblique in a few species. Mesosoma. Spines or denticles always absent from pronotum and mesonotum, present on propodeum only in M. majeri. Posterior pronotum planar, rounded or angled. Mesothoracic spiracles often opening on small dorsal prominences, these prominences stalk-like in M. hirsutus. Mesopleural process absent. Mesopleuron folds over onto flattened mesosternum without being demarcated by strong carina. Dorsal face of propodeum may be uniformly convex to flat, abruptly conical, or convex anteriad and flattened posteriad.

Propodeal angle commonly absent, but may be sharply defined and is produced as denticles in M. majeri. Declivitous face of propodeum convex to flat, longer to shorter than dorsal face. Metanotal groove usually present, forming a weak to strong angle between the mesonotum and propodeum, but is absent or vestigial in a few species. Propodeal spiracle of variable appearance, but always prominent and longer than wide, usually situated at or near propodeal declivity, but nearer to midline of propodeum in some species-groups and in individual species. Petiole. Node always present. In profile, node often squamiform, but may be thicker, e.g., subcuboidal or triangular, and is typically directed posteriad. In full-face view, dorsum of node generally rounded, more rarely angulate, weakly acuminate or indented anteromedially. Petiole with a ventral lobe or keel. Gaster. First tergite vertical and somewhat overarching the petiole in dorsal view, with a small depression for receiving the rear of the node: Legs. Mesotibia and metatibia with one simple spur in most species, spurs reduced or absent in a few groups. General characters. Integument variable, from thick and sculptured to thin, flexible and glossy in very small minor workers of some species, mesonotum may be translucent. Sculpture, where present, predominantly consisting of microreticulation or fine striation, frank rugosity or larger striae found in very few species. Mesopleuron usually with some sculpture, even in otherwise smooth and glossy species. Appressed setae often small, and well-spaced, thick in some species producing pubescence and a silvery sheen in some lights. Erect setae tend to be short and bristly, more rarely fine, long; setae modified and spatulate or even clavate distally in a few species. Many species without erect setae on mesosoma and gaster, or with erect setae confined to margins of gastral tergites. Proventriculus. Proventriculus asepalous.

Major As for minor worker, but differing in the following particulars: Head. Usually planar or weakly concave, but may be strongly concave, never strongly convex. Head often strongly horizontally rectangular, distinctly broader than wide, especially in M. wheeleri species complex. Antennal scapes shorter than in corresponding minor worker, often barely exceeding or not even attaining the posterior margin of the head. Anterior margin of clypeus very variable and may be different to that of corresponding minor worker, often folded or turned back in large-headed species, one major worker with V-shaped anteromedial protuberance projecting anteriad. Mouthparts. Palps, if reduced, as for corresponding minor worker, but may be 3,2 (less than for minor worker) in one species. Otherwise, mouthparts as for corresponding minor worker in many species, but differing significantly in M. wheeleri and M. laticeps species complexes; here, major worker with powerful, truncate, inwardly curved mandibles with four to five teeth in taxa for which major worker is known (likely to be more in unconfirmed major worker of Melophorus species K (TERC)), basal margin folded over horizontally towards buccal cavity to form a planar surface that is delimited partially or fully along its length by a carina. Masticatory margin of mandible vertical, oblique or medially indented. (Characters of mesosoma and metasoma generally as for corresponding minor worker, but dorsal profile of mesosoma may vary, with pronotum and mesonotum generally more raised and convex. Erect setae nearly always present in major worker, even where the corresponding minor worker is glabrous).