Veromessor pergandei

Veromessor pergandei is one of the commonest ants in the Mojave and Sonoran deserts. It is especially noteworthy because of (1) its glistening black color (which is unexpected in a hot dry climate); (2) its strong polymorphism, which is not shared by any of its Nevada congeners; (3) its large craters with their huge chaff piles, which are such conspicuous features of the desert landscape; and (4) its foraging column. (Wheeler and Wheeler 1986)

Identification
The examination of a large amount of material in the collection of the senior author shows that pergandei exhibits a certain degree of morphological instability. That slight differences in color and sculpture do occur has been noted in the descriptions. It is, moreover, also possible to select individuals in which the length of the epinotal spines is reduced. This last character is usually confined to the minor worker, a caste in which the spines are sometimes replaced by mere angles. (Wheeler and Creighton 1934)

Distribution
Deserts of southwestern Arizona, southern Nevada, southeastern California, western Sonora, and northern Lower California.

Distribution based on Regional Taxon Lists
Nearctic Region: United States. Neotropical Region: Mexico.

Biology
Wheeler and Wheeler (1986) :

A typical nest is in an exposed area and is surmounted by a low circular or semicircular crater of excavated soil, which is 30 cm in diameter. There is a single large (16 mm) entrance in the throat of the crater. The chaffpile - a mass of discarded husks from the harvested seeds, other rubbish and a goodly number of viable seeds-forms a crescentic or circular zone at the periphery of the crater. Well-established colonies usually have several craters, often in a cluster but sometimes scattered over an area 3-4 m in diameter. Two or more craters may be active. New craters are opened from below and lack a chaff pile. Extinct craters have the entrance closed and the chaff pile is matted down by the weather. The following account of foraging is adapted from Creighton's admirable 1953 account. In late spring and early summer V. pergandei starts its daily routine before sunrise. The workers emerge from the entrance and move very slowly (because of the low temperature) over the crater. After sunrise these movements are accelerated. Thirty to 45 minutes after sunlight first strikes the nest a foraging column pushes out from the crater and moves decisively toward the seed supply, which may be 70 or more meters from the nest. Naturally at first all workers are outward bound. Soon, however, homeward bound seed-carriers appear in the column. After 2 or 3 hours (i.e., about mid-morning) workers suddenly cease to leave the nest. The column disappears as soon as all the homeward bound workers have reached the nest.

Nest work follows. Workers emerge from the entrance each bearing a piece of gravel, which is dropped on the crater, or a seed husk, which is deposited on the chaff pile. Nest work usually continues until the second foraging period in the late afternoon. But on very hot days all ants disappear into the nest in the early afternoon. In the late afternoon nest work increases suddenly and nest workers soon cover the crater. Then out of the entrance there is a rush of workers carrying nothing, who leave the crater and start a foraging column. Foraging continues until early dusk (i.e., for about 3 hours). Nocturnal nest work may continue until early morning.

Wheeler and Creighton (1934):

Of all the species in the genus Veromessor, pergandei appears to be most completely adapted to a xerophilous mode of life. This might logically be inferred from the structural characters because of the unusual development of the ammochaetae. In this species not only are the gular and mental hairs very long and numerous but, in addition, they are augmented by another row of hairs arising along the lower lateral border of each mandible. These accessory ammochaetal hairs do not occur in the other species of Veromessor. An examination of the locality records presented above will show that pergandei occurs in areas of extreme aridity and high temperature. The fact that a number of the records come from the Mojave and Imperial Deserts needs no further comment to anyone familiar with the climatic conditions in those areas.

The habits of pergandei have been summarized by the senior author as follows: “The nests are single or more rarely multiple craters, much flattened, with rounded slopes, 50 cm. or more in diameter, and with one to three large and very irregular central openings. Sometimes these are slit-shaped and as much as 5 or 6 cm. long. The rough galleries and granaries are excavated to a depth of at least 60 cm. in the hardest and most sun-baked portions of the desert soil. Late in the afternoon long files of workers may be seen in the full activity of harvesting. Sometimes these files may be followed for a distance of 20 or 30 meters from the nest before the ants disperse among the scant vegetation in search of seeds. They seem to have no preferences, but eagerly seize all the mature seeds they find and carry them to the nest where they carefully remove the husks and store the edible kernels in the granaries. The chaff and seed-pods are then carried out and dumped on the kitchen midden, which forms a crescentric or circular zone at the periphery of the crater. Sound seeds are often thrown out with the chaff and eventually germinate, so that old nests are often marked by a circlet of growing plants. . . .”

It may be added that pergandei is one of our few desert ants that does not remain in the nest during the middle of the day in the summer months. The junior author observed it foraging at noon on the Mojave Desert in August. The temperature optimum for this insect must be exceedingly high and this point offers interesting ecological possibilities for anyone who has the temerity and constitution necessary for a mid-summer study of the species.

Nevada Wheeler and Wheeler (1986) - In winter foraging may be limited to the middle of the day. The following notes result from the month we spent studying this species at Boulder City: (1) In hot weather some foraging may be done at night, either by moonlight or in total darkness. (2) The width of one observed column increased from 2 cm to 107 cm and then narrowed to 30 cm depending upon the terrain and the plant obstructions. (3) More than one column may operate simultaneously from the same nest, or even from the same entrance. (4) No one has ever counted the population of a pergandei colony. This would probably be an impossible task, considering the rocky terrain which the species inhabits. Some idea of its enormity may be grasped from an estimate of I7,000 workers in a column 40 m long (J. Wheeler and Rissing, 1975b). (5) The longest column we have measured was 46 m. (6) We have clocked the speed of workers in a column from 5 mm/sec at 16°C to 66 mm/sec at 42°C soil surface temperature. The largest crater we have measured was at Davis Dam in Clark Co.: 2 m in diameter; a nearby crater was 1 m and there were numerous other very large craters in the vicinity. We have 48 records from 33 localities, all in the Hot Desert in the southern fourth of the state; 600-4,000 ft.

Foraging
Plowes et al (2014) - Three Nearctic seed harvesting ants in the genus Veromessor (Veromessor pergandei, Veromessor  andrei, and Veromessor  julianus) employ column foraging to guide workers leaving and returning to the nest. During summer months, V pergandei colonies foraged from ~4:30 to 8:00, depending on temperature. In winter, foraging occurred during the middle of the day when temperatures were above 13 °C. There was a predictable sequence of phases in the formation of foraging trails. This included initiation, extension, marching, harvesting, and termination. When foraging first begins for the day, tens of thousands of V. pergandei workers emerge from their colony and move along a narrow path away from the nest. These paths, the foraging column, extend from 3 to more than 40 m. At the location where the foraging column comes to an end, workers then disperse in a foraging fan, i.e., individuals leave the trail and forage independently, then return to the column after they have collected a food item (Went et al. 1972; Bernstein 1975; Wheeler and Rissing 1975). Columns may form trunk trails when they lead to a more stable resource (Plowes et al. 2013). The direction taken by a column is determined at the beginning of each foraging bout. Columns always originate at the nest and direction may change between morning and evening and between subsequent days (e.g., Clark and Comanor 1973). The columns function to direct workers to harvesting sites while simultaneously avoiding neighbors (Ryti and Case 1988; Plowes et al. 2014). Inter-colony aggression occurs when columns from neighboring colonies intersect (Wheeler and Rissing 1975; Brown and Gordon 2000).

Plowes et al. (2014) combined behavioral analyses in the laboratory and field to investigate chemical communication in the formation of foraging columns in two Nearctic seed harvesting ants, Veromessor pergandei and Veromessor andrei. Messor pergandei use poison gland secretions to lay recruitment trails. The poison gland contains 1-phenyl ethanol and minute quantities (3 μl of a 0.1 ppm solution) of this substance released trail following behavior. The recruitment effect is enhanced by adding pygidial gland secretions. Messor pergandei workers showed weak trail following to 5 ppm trails of the pyrazines 2,5-dimethylpyrazine and 2,3,5-trimethylpyrazine, and minute quantities of pyrazines were detected in their poison glands.

Freas et al. (2019) - During the fall and spring seasons, V. pergandei forages diurnally, beginning in the early morning to late afternoon with a period of decreased activity during midday. Workers forage using a combination of group and individual navigational cues. Foragers leave the nest along a pheromone-marked column (30–50 cm wide) before fanning out to forage individually (Plowes 2012; Plowes et al. 2013, 2014). Once a forager finds food, they return to the end of the column (referred to as the column head), relying on their celestial-based path integrator to find this location (Plowes et al. 2019). Once the forager re-enters the foraging column it follows the column back to the nest entrance. Displacement work in V. pergandei shows that foragers use a path integrator to return to the column head rather than the nest entrance (Plowes et al. 2019). Some evidence has also suggested that foragers track celestial-based directional cues and distance estimates as a backup strategy when foragers are displaced off the pheromone column (Freas et al. 2019a; Plowes et al. 2019), but the active use of these cues while in the column remains unknown.

Freas et al. (2019) examined how polarized light could be used by V. pergandei foragers to find the end of the pheromone trail that is used to leave from and return to their nest. Previous studies of navigation using polarized light have focused on solitary foraging species and not species that use pheromone trails. During the testing period unmanipulated colonies formed their foraging column at sunrise and stopped their outgoing trips at sunset. All the foragers remaining in the field returned to the nest via the column within a short period (20 min) after sunset. In this study inbound and outbound foragers were presented with manipulated overhead polarization patterns, and with the sun's position experimentally blocked in the evening. The results showed that foragers actively orient to an altered overhead polarisation pattern, both while navigating during their individual foraging forays and while traversing the pheromone-based column. Heading changes occurred during twilight, as well as in the early morning and late afternoon before sunset. Differences in shift size indicate that foragers attend to both the polarization pattern and, when available, the sun’s position. During twilight, headings are dominated by the polarisation pattern. Finally, when the sun’s position was experimentally blocked before sunset, shift sizes increased similar to twilight testing. These findings show that celestial cues provide directional information on the pheromone trail.

Spider Predation
Hale et al (2018) - In October 2017 in Granite Cove within the Sweeney Granite Mountains Desert Research Center, California, U.S.A. (34.7810° N, 115.6508° W), we observed approximately ten Euryopis californica spiders attacking ant workers near a Veromessor pergandei colony shortly after dusk. At this time of the evening, there were relatively few workers outside the nest, and they were slow-moving due to the decreasing temperature. The spiders foraged within 50 cm of the nest entrance, including near the entrance itself, on the midden, and just outside the midden ring. We observed several successful and unsuccessful attacks by spiders on workers. In at least four instances, E. californica individuals approached V. pergandei workers while we observed. In one case, when an ant responded defensively by opening its mandibles and facing the spider, the spider retreated. Three successful attacks involved spiders ambushing workers and rapidly wrapping them in silk. In one case, the restrained ant was subsequently envenomated through a bite to the gaster (Fig. 1A). Interestingly, while one spider envenomated the ant, two others were observed nearby (within ~5 cm). In two instances, we witnessed E. californica individuals dragging silk-wrapped ants away from the nest entrance using their spinnerets. Initially, it was unclear whether the spiders were intentionally moving dead or moribund workers, or if the silk-wrapped ants were unintentionally stuck to the spiders as they departed the nest area. To test this, we disturbed one spider that was carrying an ant by picking it up and dropping it from 10 cm above the ground. The spider dropped the ant in the fall but immediately reoriented itself, reattached the silk-wrapped ant to its spinnerets, and continued its original trajectory. This behavior suggests that the spiders attack and kill workers at the nest and move them to another location for feeding or storage. In addition to witnessing envenomation and transport of prey away from the colony first-hand, we observed two silk-wrapped workers that appeared to be moribund or freshly killed on and near the midden. These were not being actively moved or attacked by spiders, suggesting that the spiders might kill multiple ants in one hunting bout. We collected five spiders from around the ant nest, including the individual witnessed envenomating an ant, for identification. All individuals collected were male.

Kwapich and Holldobeler (2019) found that workers of Veromessor pergandei are able to rescue workers ensnared in webs of spiders (Steatoda sp. and Asagena sp.) that build webs in their foraging paths. The ants will also dismantle the web in order to remove the immediate threat of additional workers being captured. This is the second reported instance of rescue behavior being exhibited in ants (see Megaponera analis) but the dismantling of a spider web is a novel behavior for ants. More generally, prey species rarely seek out and dismantle traps constructed by their predators hence this behavior is all the more surprising and novel.

Nomenclature

 *  pergandei. Aphaenogaster pergandei Mayr, 1886d: 448 (w.) U.S.A.
 * Wheeler, W.M. & Creighton, 1934: 376 (s.q.m.); Wheeler, G.C. & Wheeler, J. 1972b: 240 (l.).
 * Combination in Stenamma (Messor): Emery, 1895c: 307.
 * Combination in Novomessor: Emery, 1915d: 73.
 * Combination in N. (Veromessor): Forel, 1917: 234.
 * Combination in Veromessor: Wheeler, W.M. & Creighton, 1934: 374.
 * Combination in Messor: Bolton, 1982: 341.
 * Combination in Veromessor: Ward et al., 2014: 13.
 * See also: Smith, D.R. 1979: 1365.

Worker
Wheeler and Creighton (1934) – Major. Length 6 mm.

Head, exclusive of the mandibles, as broad as long, widest just behind the eyes and narrowing from this point to the insertion of the mandibles. Anterior to the eye the side of the head is straight, posterior to the eye it is convex, passing through a broadly rounded angle to the straight occipital border. Clypeus scarcely projecting, its anterior edge feebly arcuate, the median lobe with a prominent, blunt central tooth from which several well-marked rugae extend rearward between the antennal lobes. Antennal lobes parallel, not spreading laterally in the rear, the frontal area between them subcircular and opaque. The lateral border of the clypeus and the anterior border of the side of the head together form a very distinct semicircular impression at the base of the mandible. Mandibles massive, their outer border strongly and evenly convex, the masticatory margin feebly concave, armed with two stout teeth at its junction with the outer border and a single smaller tooth at the junction with the inner border, the remainder of the masticatory margin feebly denticulate or smooth. Antennal scapes slender, their tips in repose just reaching the occipital border. Funicular joints all of approximately the same length and all longer than broad, their diameter gradually increasing apically. The eyes, which are set at the middle of the sides of the head, are large and elongate, with the convex upper orbit meeting the straight lower orbit in a bluntly rounded anterior angle.

Promesonotum, seen from above, subpyriform, the pro mesonotal suture obsolete above, but its position indicated laterally by a broad impression. Mesoepinotal suture with a broad and fairly deep impression. Epinotum sub triangular, slightly longer than broad but not as long as the promesonotum, armed posteriorly with a pair of feebly divergent, short, slender spines. Seen in profile, the promesonotum is strongly gibbous and descends to the much lower epinotum through an even curve, except at the point adjacent to the mesoepinotal suture where the dorsal impression of the thorax gives it a short, steep posterior face. Dorsum of the epinotum in profile straight, except for a low anterior ridge which bounds the rear edge of the mesoepinotal impression. Epinotal spines short, only slightly elevated, and arming the sharp angle between the basal and declivous faces of the epinotum.

Anterior peduncle of the petiole, seen in profile, thick, its upper face meeting the anterior face of the node in such a wide angle that the two form what is virtually a single slightly concave slope. Lower surface of the peduncle straight or very feebly sinuate, with a small angular tooth at its anterior end. Posterior peduncle very short and thick. Node of the petiole small, its height not equal to the greatest thickness of the anterior peduncle, the summit flattened and sloping rearward to the very short posterior face. Postpetiole in profile lower than the node of the petiole, with its upper surface consisting of a strongly sloping anterior face and a somewhat less sloping posterior face, which meet at the summit in a broadly rounded angle. Ventral face of the postpetiole sinuate. Seen from above, the node of the petiole is scarcely wider than the peduncle and only slightly more than half as wide as the trapezoidal postpetiole. Gaster oval, its anterior edge slightly flattened at either side of its attachment to the postpetiole.

The entire insect shining. Mandibles with numerous coarse longitudinal striae. Head with numerous, small piligerous punctures. Sides of front behind the clypeus with feeble longitudinal striae which extend rearward to the middle of the head. Antennal lobes with similar striae which spread fanwise on the front, disappearing at the level of the posterior edge of the eyes. The remainder of the head may be highly glabrous or more or less covered with coarse shagreening, heavy enough partially to obliterate the shining character of the surface. Pronotum feebly shagreened, moderately shining. Mesonotum, epinotum, and petiolar nodes finely granulate and feebly shining. Gaster very smooth and shining, with moderately numerous, small, piligerous punctures.

Hairs whitish or yellowish, slender but with rather blunt tips. Those on the under surface of the head and the lower lateral edges of the mandibles very long and numerous, forming well-developed ammo chaetae. Hairs on the upper surface of the mandibles very short and subappressed. Clypeal border with numerous long curved hairs. Those on the dorsal surface of the head sparse and of variable length but averaging longer than those of the thorax and pedicellar nodes. Gaster with moderately numerous short hairs all of approximately the same length. Legs clothed with short whitish hairs which are more numerous on the tibiae and tarsi than on the femora. Scapes covered with exceedingly fine hairs which grade into golden pubescence on the apical funicular joints.

Color piceous brown to jet black, the mandibles, antennae, and legs somewhat lighter than the remainder of the insect.

Minor Length 2.5 mm.

Differs from the major in the following characteristics: the head is slightly longer than broad, with the occipital border feebly convex in the middle. The antennal scapes in repose surpass the occipital border by an amount slightly in excess of their greatest diameter, the impression at the mesocpinotal suture is relatively broader and shallower, the sculpture is feebler, the pilosity, especially that of the gaster, is much sparser.

Medias vary between the extremes described above, depending upon their size.

Queen
Wheeler and Creighton (1934) – Length 10 mm.

Head, mandibles excluded, quadrate, slightly broader than long, the straight sides slightly narrower in front of the eyes than behind them, the occipital angles very broadly rounded, the middle of the occipital border flat. Mandibles large, their outer margin less strongly curved than in the major worker. Apex of the masticatory margin armed with two long, very stout and sharp terminal teeth, the outer tooth stouter and longer than the inner, but both far larger than the row of short, triangular teeth which arm the rest of the masticatory margin. Clypeus short, its anterior margin feebly projecting and sinuate in the middle. Median lobe with a single projecting tooth as in the major worker. Frontal carinae short and rather flat, not bounding the antennal fossae in the rear, the frontal area between them arrow-shaped, only a little depressed, and feebly shining. Antennal scapes stout, their diameter gradually increasing from base to apex, the tips only slightly turned outward. In repose the scape surpasses the occipital margin by an amount slightly in excess of its greatest diameter. All funicular joints longer than broad, their diameter gradually increasing from base to apex, the first and last joints slightly longer than the rest. Eyes large, moderately convex and reniform in outline, set somewhat behind the middle of the side of the head, with their long axes slanting forward and downward. Ocelli well-developed, the median ocellus varying considerably in size and shape throughout the type series. In some specimens it is round and no larger than the lateral ocelli, in others it is oval and notably larger.

Thorax bulky, oval when seen from above; its greatest width, which occurs behind the middle of the scutum, slightly more than half its length. Scutum large, its evenly and strongly convex anterior edge completely obscuring all but the projecting neck of the pronotum when the thorax is viewed from above. Paraptera poorly developed and apparently fused to the adjacent portions of the scutum and scutellum; the former sutures being indicated merely by impressions at either side of the paraptera. Scutellum subtrapezoidal with the angles rounded, largely concealing the metanotum when seen from above. Epinotal spines seen from above short and stout, with rather bluntly rounded tips. Seen from the side, the almost vertical anterior face of the pronotum is overhung by the bulging forward edge of the scutum. The greater part of the scutum and the scutellum together form a straight dorsum to the thorax. Mesothoracic sternite separated from the epimeron by a well-marked impression. Basal and declivous faces of the epinotum not sharply separated, together forming a single steep and slightly convex slope with the two bluntly triangular epinotal spines at the middle.

Peduncle of the petiole very thick, with a prominent anteroventral tooth. Seen in profile, its dorsal face forms an unbroken slope with the anterior face of the node. The latter is small, triangular, and with a narrow though blunt summit. Posterior peduncle of petiole exceedingly short and thick. Postpetiole cup-shaped in profile, larger than the petiole, with a very short anterior peduncle and no posterior peduncle, the entire, wide, posterior face being applied to the surface of the gaster. Seen from above, the node of the petiole is only about three-quarters as wide as the subcuneiform peduncle below it. Postpetiole subtrapezoidal, with the lateral faces slightly convex, its greatest width two-fifths again as wide as the petiole. Gaster long, sub oval, with a flattened anterior border. Femora not incrassated. Wings tinged with yellow, the veins and stigma brownish yellow.

Mandibles shining, with strong, subparallel rugae. Portion of the c1ypeus adjacent to the median -tooth with longitudinal rugae, the lateral portions finely granulose, the granulation being replaced at the extreme ends by striae which are carried backwards onto the front and curving around the lateral edges of the antennal fossae. Frontal carinae with longitudinal striae, which spread fan-wise on the vertex but do not reach the occiput. The entire head covered with small, piligerous punctures. These are most numerous on the vertex and occiput where they tend to replace the longitudinal striae which become very feeble on the posterior parts of the head. On the front the punctures are sparser and at the anterior edge of the head, where the striae are heavy, they disappear altogether. Thorax in large part smooth and more shining than the head. The scutum and scutellum with scattered punctures smaller than those on the head. Lateral portions of the pronotum, the epimeron and the upper pan of the mesothoracic sternite coriaceous, with very feeble striae impressed upon the coriaceous marking so that these parts -appear striate or coriaceous, depending upon the angle of illumination. Lower portion of the mesothoracic sternite and epinotum with coarse, wavy, longitudinal rugae, the interrugal spaces finely granulose and less shining than the remainder of the thorax. Petiolar nodes coriaceous, scarcely shining. Gaster strongly shining, feebly shagreened, with scattered small puneture.

Pilosity as in the major worker except for the gaster, on which the hairs are less uniform, those at the edge of each segment being longer, thicker, and more numerous than elsewhere.

Color piceous black, the mandibles, anterior border of the head, funiculi, and tarsal joints clear, reddish brown; the scapes, tibiae, femora, and abdomen piceous brown.

Male
Wheeler and Creighton (1934) – Length 8.5 mm.

Head small, quadrate, notably wider behind the eyes than in front of them, the feebly convex occipital border passing through much-rounded occipital angles to the very short part of the sides which lies behind the eyes. Mandibles as in the female, except that they are relatively much less massive and somewhat narrower. Anterior border of clypeus feebly convex, with a broad median impression, the median lobe projecting but without the central tooth characteristic of both major worker and female. Eyes large, suboval in outline, with their anterior border carried well under the genae. The antennal scapes in repose reach the lateral ocelli. All funicular joints longer than broad, their diameter gradually increasing from base to apex of the funiculus. First funicular joint subcampanulate or trumpet-shaped, notably broader than the adjacent joint although not equaling it in length. Funicular joints 2-6 as broad basally as apically, joints 7-11 subovoidal, with the basal ends somewhat constricted, the terminal joint rather sharply pointed and as long as the two preceding joints together.

Thorax bulky. In profile the scutum shows a fiat posterior half and a strongly convex anterior portion which slightly overhangs the pronotum. Scutellum in profile with its anterior edge slightly higher than the adjacent portion of the scutum from which it is separated by a broad suture. Posterior edge of the scutellum overhanging the very short, strap-like metanotum. Basal face of the epinotum longer than the declivous face, the angle between them well-marked and armed with a broad, short, somewhat rounded tooth at either side. Seen from above the thorax is suboval, the anterior border evenly convex, the declivity of the epinotum forming a fiat, truncated posterior face. The greatest width of the thorax occurs at the posterior third of the scutum. Mesothoracic paraptera virtually obliterated, their place being taken by a broadly impressed suture between the scutum and the scutellum. Notauli absent.

Nodes of pedicel as in the female, except that the posterior face of the postpetiole is even more broadly applied to the anterior surface of the first abdominal segment without any constriction between the two. In profile this gives the postpetiole the appearance of being a narrowed portion of the gaster rather than a separate node.

Head and mandibles finely striate-granulose and feebly shining. Median lobe of clypeus irregularly rugose, in some specimens with a median crest extending back onto the front. Dorsum of the thorax shining, the sculpture coarser but notably sparser so that there are large glabrous areas, particularly in the middle of the scutum. Sides and lower portions of the thorax feebly shining or opaque, densely striate-granulate. Petiolar nodes finely granulose, opaque. Abdomen strongly shining, rather strongly shagreened, and with numerous small punctures.

Hairs numerous, golden, erect or sub erect, curved, fairly long, and of about equal length on head, thorax and abdomen. Those on the gulae and mentum form a small but distinct psammophore. Hairs on the legs short, fine, and sub erect. Those on the antennae finer and appressed.

Color piceous black, the apical third of the mandibles and all of the tarsal joints castaneous brown, the funiculi castaneous basally but becoming clear, golden yellow toward the apex. Wings as in the female.