Solenopsis enigmatica

A rare parasitic species known from type material collected from within two different nests of Pheidole antillana. It has only been sampled from the Caribbean island of Dominica.

Identification
Deyrup & Prusak (2008) - Worker. Distinguished from other Solenopsis workers by clypeal morphology: clypeus lacking carinae, in lateral view rounded and strongly declivitous. Pale yellow, antennal scapes short, in frontal view failing to reach occipital margin of head by a distance about equal to the combined length of the first 4 segments of the funicle. If S. phoretica produces workers, they might be similar to those of enigmatica, so this diagnosis may not be definitive.

Dealate Female. Distinguished from most other Solenopsis by the form of the mandibles, which are falcate, with the teeth not in an oblique series, as normal in species of Solenopsis; the basal angle is strongly developed and weakly bidentate, the concave inner margin with a low, median double tooth; Solenopsis phoretica lacks this median double tooth.

Distribution
Deyrup & Prusak (2008) - All specimens were collected in Dominica rain forest habitat as defined by Lack et al. (1998), with evergreen shrubs, canopy trees and abundant epiphytes.

Distribution based on Regional Taxon Lists
Neotropical Region: Dominica.

Biology
Deyrup & Prusak (2008) - There is a striking resemblance between the queens of enigmatica and Solenopsis phoretica. With only one specimen of phoretica and two of enigmatica, little can be said about the variation in either species. It seems unlikely, however, that these specimens represent a single, widespread, variable species. The differences in mandibular structure are probably functionally significant, considering that the specimen of phoretica was found with her mandibles locked around the petiolar peduncle of the host queen. Inquiline ants tend to be host-specific, exploiting a single host, or a few very closely related hosts (Hölldobler & Wilson 1990). The S. phoretica-enigmatica group is known from only 3 queens and a few associated workers, but all 3 were apparent inquilines in the nest of Pheidole species, and we are assuming that this is a normal association. It is rare for an inquiline ant to be so distantly related to its host, but there is a relevant example in Oxyepoecus, a genus related to Solenopsis, with hosts in both Pheidole and Solenopsis (Hölldobler & Wilson 1990). We are not suggesting that there is any reason to think that phoretica and enigmatica could be transferred to Oxyepoecus, or that they represent some sort of transition between Solenopsis and Oxyepoecus. Several species of small Solenopsis may live in nests adjacent to those of other ants, which they exploit in what one might call a periinquiline way (Hölldobler & Wilson 1990), and it could be that the phoretica-enigmatica lineage took such a relationship a few steps further.

The exact nature of the relationship between S. enigmatica and its host is unknown. It is reasonable to speculate that queen enigmatica spend some time in close association with the host queen, as enigmatica has the elongate mandibles and concave first gastral tergite found in phoretica. No queen Pheidole was found in either of the 2 colonies that contained enigmatica, but it is possible that the host queens were overlooked. The presence of 3 workers associated with 1 queen enigmatica is consistent with the hypothesis that enigmatica is a temporary nest parasite, but no mature colony has been found. Worker enigmatica are conspicuously larger than the other 3 species of yellow Solenopsis that we found on Dominica, and it is unlikely that we would have discarded a colony of enigmatica, mistaking it for one of the common little yellow Solenopsis. Parasitic ants in general are notoriously rare. The abundance of another parasitic Solenopsis, Solenopsis daguerrei has been studied by Calcaterra et al. (2000). With 4,131 potential host colonies examined, parasitization rates were 3.9%. If this degree of rarity is at all typical of parasitic Solenopsis, we were fortunate to find 2 parasitized colonies during the short time we were in Dominica.

It is highly unlikely that the phoretica-enigmatica lineage of Solenopis is confined to the island of Dominica and Alachua County in Florida. One possibility is that members of this group occur throughout the range of Pheidole in the New World, but have escaped notice because they are rare and host-specific. It is also possible that this group of inquilines can only succeed in communities in which a few species of Pheidole are especially abundant, with persistent and accessible colonies. This might favor the success of inquilines on islands with relatively low Pheidole diversity, and near the edges of the range of Pheidole. The relentless raiding of Pheidole nests by army ants in the humid Neotropics would seem to decrease the effective resource base for inquilines.

The potential host range for S. enigmatica on Dominica is probably limited. The species most similar to P. antillana is Pheidole laudatana Wilson, a slightly smaller member of the tristis group that is common on Dominica, but not known from other islands. Both species are common in primary forest on Dominica. Pheidole antillana nests in rotten sticks and logs on the forest floor, while colonies of laudatana nest in clay soil and have a small nest opening. It seems probable that antillana is more vulnerable to inquilines than laudatana. If it turns out that S. enigmatica is a host-specific inquiline, it must have a restricted distribution, as P. antillana is known only from St. Vincent, Grenada and Dominica. We found antillana common in the extensive forest reserves on Dominica, but absent from disturbed areas. Other islands of the Lesser Antilles have less extensive forest reserves, in some cases only small fragments. It is too soon too speculate about the rarity of S. enigmatica, but its conservation status is one of several interesting questions pertaining to the species.

Nomenclature

 *  enigmatica. Solenopsis enigmatica Deyrup & Prusak, 2008: 71, figs. 1-2 (w.q.) DOMINICA.

Worker
Paratype. Total length (length of head excluding mandibles + length of mesosoma + length of petiole + length of postpetiole + length of gaster): 2.27; head length: 0.50; head width at rear margins of eyes in frontal view: 0.45; length of mesosoma: 0.60; length of petiole: 0.20; length of postpetiole: 0.12; length of gaster: 1.55. Color: pale yellow, eyes black. Head: smooth, shining, with scattered setigerous punctures, including a series of 4 large, posteriorly-directed, suberect, slightly convergent setae on each side of frontal area, and several small, decumbent, convergent setae on each side of frontal area; eye composed of a single, large, dark facet; mandibles unlike those of queen, more typical of workers of other small Solenopsis species: apical tooth long and sharp, remaining 3 teeth small, sharp, in a strongly oblique series, with a long, toothless proximal inner margin; clypeus without carinae, convex in lateral view, with four subapical elongate setae and a long median apical seta; antennae 10-segmented, scapes in frontal view failing to reach occipital margin by a distance about equal to the first 4 segments of the funicle; antennal club 2-segmented. Mesosoma smooth, shining, pronotum with 3 large, semierect, posteriorly-directed setae on each side, and 1 smaller, semierect, anteriorly-directed seta on each side at posterior margin of pronotum; propodeal spiracle large; legs smooth, shining, with sparse, strong, semidecumbent, distally- directed hairs; hind basitarsus about the length of hind tibia. Petiole: apex of node in posterior view somewhat flattened; node with two long, posteriorly-directed setae on each side near apex; ventral process a low, triangular tooth, somewhat rounded apically. Postpetiole: low and rounded above in lateral view; in dorsal view 5/7 as long as wide. Gaster: first tergite with sparse, long, suberect, posteriorly-directed hairs that are generally distributed; tergites 2-4 smooth each with an irregular row of large, suberect, subapical hairs; sting slightly extruded.

Queen
Holotype Dealate. Features visible in lateral view, described from left side, mm. Total length (length of head excluding mandibles + length of mesosoma + length of petiole + length of postpetiole + length of gaster): 3.92; head length: 0.60; head width at rear margins of eyes in frontal view: 0.62; length of mesosoma: 1.20; length of petiole: 0.37; length of postpetiole: 0.20; length of gaster: 1.55. Color: yellowish brown, ocellar triangle dark brown, appendages yellow. Head: smooth, shining, sparsely covered with setigerous punctures separated by 2-8 times the width of a puncture in frontal area, 2-4 times the width of a puncture on sides, setae suberect, directed posteriorly in frontal area, elsewhere directed anteriorly; ocelli not enlarged, each ocellus about the width of antennal scape near base; malar distance about half (4/7) length of eye; mandibles elongate, a little less than half the length of the head at midline; mandible with basal angle strongly developed, bidentate; inner margin of mandible strongly concave distal to basal angle, with a low, median, marginal double tooth, the proximal cusp blunt but conspicuous, the distal cusp reduced to a low ridge; clypeus declivitous, concave, without carinae, with 4 subapical elongate setae; antennae 10-segmented, scapes reaching outer corner of head in frontal view, antennal club 2-segmented, club about as long as remainder of funiculus. Mesosoma: smooth and shining, with sparse setigerous punctures on pronotum and near margins of mesonotum, mesopleura, sides of propodeum; katepisternum with sparse setigerous punctures; disc of mesonotum and declivity of propodeum not punctate; propodeum evenly declivitous in lateral view; legs smooth, shining, with sparse, strong, semidecumbent, curved, distally directed hairs; hind basitarsus longer than hind tibia; a dense, coarse, ventral brush of hairs on distal half of front basitarsus. Petiole: peduncle short, less than 0.25 length of base of petiole in lateral view; node of petiole in lateral view triangular, apex broadly and smoothly rounded, in posterior view semicircular; ventral process narrowly expanded, with a small ventral projection that is like a low ridge rather than a tooth. Postpetiole: low and rounded above in lateral view, in posterior view about 1.5 times as wide as long. Gaster: in dorsal view first tergite with prominent, rounded anterior corners, projecting forward from midline; first gastral tergite and sternite each with a broad, median, basal concavity; first tergite with sparse, long, suberect, posteriorly-directed, hairs that are generally distributed; tergites 2-4 smooth, each with an irregular row of large, suberect, subapical hairs.

Type Material
We have designated a dealate queen as the holotype because of the apparent close relationship between this species and Solenopsis phoretica. Diagnosis of the new species must be by reference to character states of queen S. phoretica, as workers of that species are unknown.

Collecting data on labels (3) of holotype: Dominica, W.I., Pont Casse, 30 April 2006, M., N., L. Deyrup; Trail to Trois Pitons, in nest of Pheidole antillana in rotten log; rainforest near base of trail, N15°22.80’, W61°20.46’.

Additional Paratype Material. Two workers, one a pharate adult in pupal skin, associated with the holotype female. One dealate queen: collecting data as on labels: Dominica, W.I., Cochrane, west of Morne Macaque, 29 April 2006. Z. Prusak. Middleham Falls Trail, in nest of Pheidole antillana in rotten log. 15°20.852’N, 61°20.698’W.

Holotype and other type material to be deposited in the Museum of Comparative Zoology, Harvard University, Cambridge, MA.

Etymology
From aenigma (Latin), meaning “mystery,” referring to the many unknown aspects of the natural history of this unusual Solenopsis.