DESCRIPTION OF THE ERGATOID QUEEN OF POGONOMYRMEX MA YRI WITH NOTES ON THE WORKER AND MALE ( HYM . , FORMICIDAE )

During a recent stay in Santa Marta on the north coast of Colotnbia, I had the opportunity to study Pogonomyrmex mayri, the sole tnetnber of the subgenus Forelomyrmex, whose entire range is the desert and dry deciduous forest below 200 tn. on the northwestern, western, and possibly southern skirts of the Sierra Nevada de Santa Marta. This ant was described by Forel (1899: 61-62, footnote) from worker(s) and tnale(s) he collected. Neither he nor subsequent entotnologists, including P. J. Darlington, found fetnales. The reason fetnales were unknown became clear as I worked in the area and later began to look at the biology of P. mayri tnore closely. Though I collected tnales frotn vegetation nearly year around (3 Sept. to 30 June), no winged fetnales were seen in two years. Furthermore, only after thoroughly excavating 10 nests were any fetnales found at all, one in each of 2 nests dug by tny coworker, Maria del Cartnen Hincapi6, and her assistant. Both were ergatoid nest queens. This paper presents a fortnal description of those queens, notes on the worker and male suppletnenting Forel’s original description, and a discussion of the taxonotnic status of Pogonomyrmex mayri. Notes on the biology of P. mayri will be reported later.


INTRODUCTION
During a recent stay in Santa Marta on the north coast of Colotnbia, I had the opportunity to study Pogonomyrmex mayri, the sole tnetnber of the subgenus Forelomyrmex, whose entire range is the desert and dry deciduous forest below 200 tn. on the northwestern, western, and possibly southern skirts of the Sierra Nevada de Santa Marta. This ant was described by Forel (1899: 61-62, footnote) from worker(s) and tnale(s) he collected. Neither he nor subsequent entotnologists, including P. J. Darlington, found fetnales. The reason fetnales were unknown became clear as I worked in the area and later began to look at the biology of P. mayri tnore closely.
Though I collected tnales frotn vegetation nearly year around (3 Sept. to 30 June), no winged fetnales were seen in two years.
Furthermore, only after thoroughly excavating 10 nests were any fetnales found at all, one in each of 2 nests dug by tny coworker, Maria del Cartnen Hincapi6, and her assistant. Both were ergatoid nest queens. This paper presents a fortnal description of those queens, notes on the worker and male suppletnenting Forel's original description, and a discussion of the taxonotnic status of Pogonomyrmex mayri. Notes on the biology of P. mayri will be reported later.  22, 1978 (#780422-6)and April 26, 1978 (#780426-9). The following description is based prinarily on the first specimen, with occasional remarks on the second if they differ in sone way. For each measurement, that of the first specimen is followed by that of the second.
Mandibles subtriangular; basal and tnasticatory borders rneet at obtuse angle; outer border gently convex, slightly flattened midlength; nasticatory border with 5 teeth increasing in size apically, the second tooth with a denticle on its distal edge (second specimen with denticle reduced on one mandible, absent on the other). Palpal formula 3,3 as determined frotn undissected specinens; neither palp reaching posterior rnargin of buccal cavity. Labrun with rounded lobes on either side of a sharp median ernargination. Head ( fig. 4) about as wide as long; in full face dorsal view occiput broadly Vshaped, with distinct occipital lobes; sides of head gently convex on temples to small bulge behind the eye, then indented to flat cheeks. Middle of clypeus weakly convex overall, both transversely and longitudinally; sculpture stops a weak rnedian carina short of reaching apron; apron narrow, leading edge convex with nedian tooth bracketed by low rounded lobes; lateral artns form low narrowly rounded ridges in front of antennal fossae. Frontal triangle reduced to a narrow, deep Y-shaped sulcus separating clypeus from posterior 3A of frontal lobes. Frontal lobes short, well separated; carinae convex. Eyes small, convex, ringed by a roundbottom sculptured groove.
Antennae 12-rnerous. Scape short and rather thick, not reaching vertex of head; bent at a nearly right angle within basal 1/4 of length and largely flat beyond (extensor surface viewed from side). First 2 segnents of flagellum longer than wide, L 0.21, 0.20; 0.18, 0.18; segments 3-7 wider than long; segments 8-11 about as long as wide. Apical segtnent tnuch longer than any other, L 0.34, 0.37; apex narrowly rounded. Last 4 segments constitute a weakly defined Trunk in full dorsal view ( fig. 1) with complete, fine, itnrnobile, prornesonotal suture; tnesonotutn bisected at about 2/3 of length by a faint shallow sulcus (probable rernnant of scutoscutellar suture) which drops out at lateral rnargins; metanotutn reduced to a sharp groove between rnesonoturn and propodeurn. Propodeal spines robust, acute, their bases widely separated by a broad concavity. Propodeal spiracles prorninent; their orifices circular, facing posterolaterad.
Petiole robust (figs. 1-3), with broadly conical anterior peduncle and weakly differentiated posterior peduncle. Venter seen frorn below with 3 longitudinal ridges running frorn short acute tooth to posterior peduncle. Node in side view with flat anterior face (except for 2 low undulations) joining anterior peduncle at about a 135 angle, meeting dorsal face of node at an acute angle; dorsal face fiat, broadly curving into short posterior face. Outline of node frotn above ovoid, slightly wider than long, widest just caudad of rnidlength. Anterior edge sharp (with low upturned burr in second specirnen), gradually softening to broadly rounded along sides.
From directly above (not as in fig. 1) node subtrapezoidal; posterior edge a broad convex arc somewhat flattened mesad, posterior corners broadly rounded, sides flat, tapered to corners of weakly concave anterior edge.
Most of body with fine dense punctulate striae overlain by a coarse broken wavy sculpture (figs. 1-4): clypeus, top, sides and back of head, trunk (except those areas listed below), petiolar node and postpetiole. In other places the overlying sculpture is tnuch reduced or absent, leaving largely flat surfaces of fine, punctulate striae: gula (sotne weakly undulate sculpture present), anterior face of pronotutn, prosternutn, forecoxae, dorsotnedial surfaces of tnetapleural gland bullae, declivous face of propodeutn (striae coarser, less dense), most of first gastric segtnent (striae weaken caudad). Striation tnuch reduced, leaving predotninately fine, densely punctate surfaces on: antennae (striae present but finer than elsewhere and not predotninate), legs except first coxae, and end of first and tnost of succeeding segtnents of gaster (narrow coriaceous tnargins on each). Mandibles finely longtitudinally striate, without punctures. Inner surface of tnandibles, peduncle and venter of petiole stnooth and shining; the latter with scattered punctures. Striation tnore or less longitudinal on tnandibles, antennae, clypeus, dorsutn and sides of head, tnesonotutn, sides of trunk (though tnuch confused in parts), coxae, anterior face and sides of petiolar node, sides and venter of postpetiole, and gaster. Striation essentially transverse on gula (bisected by median longitudinal ridge), pronotutn, basal and declivous faces of propodeurn (striae converge on apices of spines frotn all sides), dorsal surface of petiolar node, anterior and dorsal faces of postpetiole (transversely arcuate).
Whole body, except peduncle of petiole, covered with short, stiff, erect, acute golden hairs, interspersed with shorter, tnore flexuous recurved hairs. Apron of clypeus with longer flexuous hairs; hair on tnandibles tnore decurnbent. No psatntnophore of any sort. Color unifortnly dark ferruginous brown, except for brown to yellow apical antennotnere.

WORKER (Figs. 6-8)
The worker is nost strikingly different frotn the queen in its jet black color, its larger head, and stnaller waist and gaster. It appears as if there is a longitudinal gradient of allornetric growth in the queen such that the anterior half grows slower than, and the posterior half faster than in the worker. A detailed description of how the worker differs from the queen follows (see also table 1).
Head much larger, more elongate ( fig. 6) 5). Clypeus rather long, broadly convex in both dimensions, middle subsiding evenly to sides; free margin broadly arcuate except for slight median flattening or ernargination, which possesses in sotne specitnens a very stnall acute or rounded tooth that is a continuation of a weak tnedian carina running the length of the clypeus. Frontal triangle triangular, with a broad, shallow, V-shaped itnpression. Frontal carinae reduced to rims encircling the sockets of the antennae. Palpal fortnula 4, 3. Scape tnuch shorter than cotnbined length of Table 1. Measurements and indices of Pogonomyrmex mayri females and workers. Data for the first female specimen are listed first, followed by those of the second. Worker data are ranges from 6 individuals selected to include the largest and smallest available from 7 collections. Measurements are expressed in millimeters; indices in percentages, and both follow the standard definitions (see Brown, 1953: 11-14;1975: 3-4). Head width was measured just behind the eyes. Postpetiolar and gaster measurements were taken separately, and from dorsal view. second and third flagellar segtnents. Scape and flagellutn fairly densely covered with short erect hairs; segments 2-12 of flagellutn also very densely endowed with fine appressed pilosity. Trunk uniquely shaped and proportioned, as if the propodeurn has grown forward, cotnpressing the rnesoscuturn and pronoturn, and rotating the neck and head to a more ventral position (fig. 5). Pronotutn very constricted rnesad. Mesonoturn very short (rnesoscutal L/WL 20-23%); seen frorn the side evenly and rather strongly convex, from above nearly equilaterally triangular; with 2 short black lines that indicate underlying apophyses of the notauli. Middle of scutellutn and rnetanoturn raised into a prominent subcircular bulge. Propodeurn elongate (propodeal L/WL 49-52%), dorsal face curving insensibly into declivous face; unarmed. Petiole long, cylindrical, nodeless. Legs long and slender (hind fernur L/WL 83-86%); front coxae long, cotnpressed front to back ( fig. 5).
Wings evenly covered with fine hairs. Venation of fore wing variable. Of 154 wings (77 individuals) from 7 collections, the most cotnrnon venation had a small closed discoidal cell with subequal sides, a hexagonal cubital cell with a narrow opening to the costal cell and sinuate lower edge (Rs), radial cell open distad, and veins Rs and M unjoined by a cross vein beyond the discoidal cell (121 wings). In 11 wings the discoidal cell was open distad, but otherwise the same. In 20 wings the discoidal cell was closed, but a cross vein (r-m) connected the Rs and M veins, creating a second cubital cell. That cross vein joined the Rs well proxirnad of the end of the first cubital cell, except in one wing where it was almost even with the end of the cell. Two wings had both open discoidal cells and the r-rn cross vein. Wings of different venation comnonly appear on the same individual.
Hypopygiurn in ventral view subtriangular; apex broadly rounded, proximal corners square with slender truncate lateral projections, middle of base with a long slender truncate anterior process. Gonostyli (=paraneres) of genital capsule tapered in side view, but with apices broadly rounded; setae occur only around apices. Digitus long, slender, blunt, strongly down-curved; not reaching to apex of gonostylus. Cuspis short, pollicate when seen from the side.
Mesoscutal and propodeal lengths measured frorn lateral view by taking their maximum length along lines parallel with Weber's length.
Aedeagus fairly slender, with serrate ventral margin; serrations decrease in size to apex. Apex narrow, blunt; half smooth, half finely serrate. Inner dorsal margins of gonocoxites fortn a long narrow, gently convex V.
Sculpture like that of worker and female, with the following exceptions: Striation on head largely transversely arcuate caudad of antennae, clypeus without coarse undulations. Trunk and petiole without overlying broken wavy sculpture; all but pronotum longitudinally striate. Striation on postpetiole gives way to purely punctate sculpture in caudal half. Gaster, first and second antennorneres, and legs smooth and shining, except for weakly striate fore coxae.

DISCUSSION
Pogonomyrmex mayri clearly belongs to the genus Pogonomyrmex as presently constituted; and is most closely related to members of the subgenus Ephebomyrmex. It is most like Ephebomyrmex in 15 of the 22 characteristics used by Cole (1968) to distinguish nales and workers of the subgenera Ephebomyrmex and Pogonomyrmex, the sting apparatus most resembles that of the Ephebomyrmex species I have examined (Kugler, 1978), and some of its most unusual characters, such as the elongate head of the nale, the Y-shaped frontal triangle of the queen and worker, and the circunocular groove, nay be seen as the extreme developnent of characteristics of Ephebomyrmex species. Nevertheless, it is remarkably different frotn any known Pogonomyrmex, with a nunber of novel characters, and consequently has been placed in its own subgenus (Forelomyrmex Wheeler) since its description. The following shows how other Pogonomyrmex species compare with mayri's most distinctive features.
The clypeus in nost Pogonomyrmex has a concave leading edge except in angustus, darlingtoni, odoratus, schmitti and townsendi (all Ephebomyrmex), none of which has a rnedian tooth. The frontal triangle is usually broadly triangulate, but is sotnewhat elongate, laterally compressed, and except for a nedian carina, depressed ZExcept where indicated, based on the MCZ collection containing 51 of the estimated 67 presently standing species, subspecies and varieties of the subgenera Ephebomyrmex and Pogonomyrmex.

Psyche
[June-September below the level of the clypeus only in darlingtoni, saucius and schmitti. In none of these, however, is the front so narrow as in mayri, nor is it at all Y-shaped. The back of the head in full dorsal view is broadly and shallowly concave in most Ephebomyrmex species, but is only excavated to such a degree that it has definite occipital lobes in the majors of some subgenus Pogonomyrmex species, e.g., badius. The cephalic index approaches that of mayri only in cunicularius (84), an undeternined species near cunicularius (84, 87), odoratus (85-87), and angustus (86-89). Only in the latter is the occiput at all concave. In sone species the sculpture is flattened at the edge of the eye, but only in darlingtoni does it becone at all impressed, and then it is shallow, only weakly defined, and lirnited to the dorsal half of the eye. All species exalnined have rnuch larger eyes, relative to the size of the head, than mayri. The petiolar node in nost Pogonomyrmex species is well rounded on top, sides and apex, and has a distinct posterior peduncle. A few species have a broad subacute to acute apex, but only in the sp. near cunicularius does the node even superficially resemble that of mayri. On closer exatnination, it also is quite distinct. Sculpture in the subgenus Ephebomyrmex tends to be "coarsely rugo-reticulate" (Cole 1968: 35), but no species exanined has the overlying broken, undulate pattern of mayri.
Ergatogyny has occasionally been reported in Pogonomyrmex, but nost specinens age only occasional aberrations in species with normal queens, e.g., comanche, maricopa, subnitidus, californicus (Cole 1968: 175), and pima (MCZ). Only one fetnale has been reported for cunicularius, and it is ergatoid (Santschi, 1931), but the description indicates nothing tnore rernarkable about it than a tnore or less distinct scutellurn. Ergatogyny seetns to be the rule in laticeps. Kusnezov (1951: 273-275) describes the range of ergatoid fortns, but tnakes no tnention of enlarged waists or gasters, or of reduced heads. Those characters are evidently unique to mayri.
The bizarre fortn of the P. mayri male is also apparently unequalled in this genus (Cole, 1968;Creighton, 1952;Gallardo, 1932;Kusnezov, 1949Kusnezov, , 1951. Sorne males of Aphaenogaster species have elongate heads constricted behind. In the other Pogonomyrmex 3Two specitnens collected by W. L. Brown in Argentina: Catatnarca, Cat. (airport), 4 Feb. 1967;Prov. Tucutnan, Ktn 1333, Rte. 9, N. of Tapia, 25 Jan. 1967 species I was able to exarnine directly, the head is at best only slightly longer than wide in sotne species (CI range of 4 Ephebomyrmex species 89-97). The tnesoscututn is long (rnesoscutal L/WL 33-50%), propodeutn short (propodeal L/WL 26-37%) and the petiole has a distinct node. The legs are shorter than in mayri (hind fetnur L/WL 61-76%), and the front wing usually has a r-rn crossvein that joins the Rs at or distal to the end of the first cubital cell (see Cole, 1968: 25-26, plate fig. 1). If the r-rn vein is absent, the first cuboidal cell is open.
Pogonomyrmex mayri rnay in fact deserve full generic status, but for the present it seems prudent not to create a tnonotypic genus before Pogonomyrmex is cotnpletely revised. The nost recent revisions (Kusnezov, 1951;Cole, 1968) have been regional in scope and thus inadequate to address the question of whether the subgenera Pogonomyrmex and Ephebomyrmex are really two distinct genera. Should such a split occur, mayri would no doubt be placed in the separate genus Forelomyrmex.