Ponera woodwardi

Known only from the mountains of central Upolu, Samoan Is.

Identification
Taylor (1967) - The worker is unique among the medium sized Ponera (those with HW 0.43-0.60 mm) by virtue of its relatively narrow head (CI 76-81) and long scapes, which clearly exceed the occipital border, and yield scape index values of 98-102. No cephalic index value less than 80 is known in any other medium sized Ponera, and only one other species, Ponera loi, has a CI range extending below 80. All other species of Ponera with scapes exceeding the occipital border have cephalic index values higher than 88, and the next lowest scape index value in the whole genus is 95.

Additional diagnostic features include the size (HW 0.53-0.60), the presence of a distinct median clypeal tooth, the somewhat enlarged eyes and the tendency for reduction of the mesometanotal suture.

P. woodwardi queens, like the workers, are characterized by their narrow heads, and relatively long scapes.

P. woodwardi is easily distinguished from most of the other Samoan Ponera species by its considerably larger size and dark coloration. P. loi has approximately the same head width as woodwardi but the head is broader, and the scape index lower.

Distribution
The known distribution of woodwardi, like that of P. loi, is limited to the Afiamalu-Malololelei area at about 650-700 m elevation on Upolu. This distribution is reflected not only in recent collections but also in those dated 1940 and 1956. All of the above collectors worked extensively at a variety of stations at all elevational levels, not only on Upolu but also on Savaii and Tutuila. Notwithstanding, woodwardi has not been taken away from Afiamalu.

Distribution based on Regional Taxon Lists
Indo-Australian Region: Samoa.

Biology
Taylor (1967) - Where present this species is not uncommon in rain forest. During my Samoan field work I often took strays under moss on logs or standing trees, or in wood fragments lying on the forest floor. Like other Samoan endemics woodwardi had apparently undergone ecological release in an area of partly cleared and rather disturbed forest at Afiamalu, and was more abundant there than in the adjacent undisturbed forests. This situation is discussed under Ponera incerta. Specimens of woodwardi were taken in berlesates of tree moss collected as high as 15 m above ground level.

The type colony was in a rotting branch fragment about 35 cm long and 8-10 cm in diameter, lying on the floor of the cleared forest area. The nest was rather diffuse, with adults and brood distributed throughout the more rotted half of the branch, in preformed crack crevices or beetle galleries. A number of nests were collected under thick moss on the upper surfaces of the many logs lying in the partially cleared forest area; and sufficient Berlese funnel collections were taken to indicate that nests are commonly found in these situations. Brood, in these cases, was located either in galleries under the moss, against the log surface, or in preformed cavities in the more rotted parts of the wood itself.

Several observation colonies were maintained for periods of 2 to 6 days. Aspects of colony behavior and brood care, including studies on the function of the larval tubercles, were carried out. Unfortunately feeding could not be induced by presentation of numerous small insects and other arthropods. On one occasion I discovered two apparent middens of small (ca 10 mm long) diplopods, close to colony nuclei of woodwardi nesting under log moss. Similar diplopods were not accepted in observation nests.

The introduced Hypoponera confinis (Roger) an ant of similar size and general ecology to woodwardi, occurs fairly abundantly at lower elevations on Upolu, where woodwardi is apparently absent. H. confinis was not encountered at Afiamalu by me but a specimen was collected by Zimmerman in 1940. While in Samoa I gained the distinct impression that the niche-equivalent of confinis at low elevations was occupied by woodwardi at Afiamalu, and consider it possible that woodwardi has contracted its range to higher elevations, through competitive inferiority to confinis in lowland areas. If this impression is correct the outcome of a competitive situation involving these two species may be influenced by ecological factors associated with elevational change, or with the degree of human disturbance in the environment. H. confinis is a fairly widespread tramp species, ranging at least from India and Ceylon through Melanesia and Polynesia to the Society Islands. It seems to have a capacity for survival in disturbed situation; a capacity perhaps less developed in woodwardi.

Nomenclature

 *  woodwardi. Ponera woodwardi Taylor, 1967a: 62, figs. 7, 55-58 (w.q.m.l.) SAMOA.

Worker
Holotype. HW 0.55 mm; HL 0.73 mm; SL 0.56 mm; CI 76; SI 102; PW 0.44 mm; WL 0.97 mm; PH 0.50 mm; PNL 0.25 mm; DPW 0.37 mm; PNI 84. Mandibles with 3 well developed teeth occupying approximately apical 2/5 of masticatory border; remainder of border with 6 or 7 irregular, medium sized denticles. Eyes moderately well developed, indistinctly 7-faceted, their diameter about 0.05 mm. Anterior border of eye located approximately 0.93 X the distance from lateral occipital border to midpoint of anterior genal border. Scapes exceeding occipital border by slightly less than their maximum thickness. Funiculus lacking a segmentally differentiated club. Anterior face of clypeus with a distinct, moderately acute, median tooth, about 2/3 as high as its width at base. Palpal formula: Maxillary 2: Labial 2 (inspected). Mesometanotal suture not present as a sharply incised line breaking sculpture, but represented by a fairly distinct, narrow, shallow indentation traversing mesosomal dorsum. Posterolateral margins of propodeum fairly narrowly rounded, forming angles of about 90°, viewed from above.

Petiolar node, in side view, massive, sub-rectangular, tapering only slightly dorsally. Dorsal surface, viewed from above, forming slightly less than a half-circle; posterior face feebly concave; posteroventral angle fairly obtuse; fenestra elliptical, small but distinct.

Mandibles and clypeus smooth and shining. Cephalic dorsum densely punctate, opaque to sub-opaque; sides of head less densely punctate, punctures smaller and surface feebly shining. Dorsum of alitrunk moderately shining, less densely punctured than head. Punctures of pronotum and propodeum finer and more spaced than those of mesonotum; those of propodeum most diffuse. Sides of alitrunk shining, with a few scattered point-punctures. Lower aspects of mesepisterna and lateral propodeal faces with a few very fine, partly indistinct, longitudinal striae. Petiole largely smooth and shining with a few scattered punctures laterally and dorsally. Postpetiole and gaster moderately punctate, their surfaces fairly shiny ; sculptural intensity diminishing posteriorly.

Moderately long erect yellowish hairs present on mandibles, clypeus, postgenae, dorsum of node, lower edge of subpetiolar process, and entire gastric surfaces. Those on gastric apex longer and coarser. A few short fine erect hairs on scapes and anterior pronotal dorsum. Pubescence adpressed, abundant on head, antennae, mesosomal dorsum and gaster. Less abundant on sides of mesosoma.

Mandibles, clypeus, frontal lobes, antennae, and legs medium yellowish brown. Head and body largely very dark brown, with a reddish brown cast on neck and edges of pronotum, lower aspects of mesepisterna and lateral and declivitous faces of propodeum, sides of petiole, subpetiolar process, ventral and apical parts of gaster, and posterior parts of its tergites.

Paratypes. 35 paratype workers from various collections made at Afiamalu between 1940 and 1962 (see below under "Material Examined") have the following dimensions and indices: HL 0.69-0.75 mm; HW 0.53-0.60 mm; SL 0.55-0.60 mm; CI 77-81; SI 98-102; PW 0.44-0.48 mm; PNL 0.23-0.25 mm; PH 0.48-0.53 mm; DPW 0.35-0.39 mm; PNI 78-84. The specimens agree well with the holotype in general structure. Number of posterior mandibular denticles varies from 5 to 7. Eyes vary in diameter from about 0.04 mm to slightly less than 0.06 mm, with 5 to 8 moderately distinct facets; scapes always clearly exceed median occipital border by a distance of up to slightly more than 2/3 their maximum thickness.

The degree of impression of the mesometanotal suture varies. Minimally developed, it is a broad shallow impression crossing the dorsum of the mesosoma. In most specimens this is fairly narrow and distinct, and even with strongly reflected light, no sharply incised sutural trace is visible. Some examples, however, have a fine incised suture, readily visible in reflected light.

The tendency for parts of the body to be lightly infuscated, relative to the general coloration, seems to be common in this species. In some specimens the extent of lighter areas is greater than that described for the holotype, while others are almost uniformly dark brown; woodwardi is evidently slow in developing full adult coloration following emergence.

Queen
Paratypes (Based on 3 queens, an alate and 2 dealates, collected at Afiamalu and Malololelei by Woodward, I.1956.) HL 0.73-0.75 mm; HW 0.60-0.61 mm; SL 0.60-0.61 mm; CI 80-84; SI 99-102; PW 0.52-0.55 mm; DPW 0.37-0.40 mm; PNL 0.24-0.26 mm; PH 0.54-0.55 mm; PNI 71-73; maximum diameter of eye 0.20 mm; ocular index 33-34. Differing from the worker in the usual characters of full sexuality and conforming to the basic plan of queen structure in Ponera. Wing venation of coarctata type.

Male
Paratypes. (Based on 2 males collected with the holotype nest series.) HL 0.55 mm; HW (across eyes) 0.53 mm; CI 96; WL 1.08 mm; PNL 0.22 mm; PH 0.32 mm; DPW 0.22 mm; maximum diameter of eye 0.27 mm; ocular index 51; palpal formula (1 specimen dissected): Maxillary 2: Labial 2, as in the queen castes. Conforming to the general structural plan of Ponera pennsylvanica. Wing venation as in queen. Structure of terminalia of the same basic pattern as in P. pennsylvanica; pygidial spine relatively strong; dorsal process of paramere broader and less digitate, lacking thickened struts along its edges.

Head and mesosoma very dark blackish brown, metasoma slightly paler; antennae, legs, wing veins, and genitalia paler-dull medium brown. The male of P. woodwardi has a relatively low cephalic index value compared with other similar sized Ponera, known males of which all have CI values exceeding 100.

Type Material
Holotype worker and paratypes of all castes in collection. Paratypes, including males and queens, in BISHOP, and. Worker paratypes in the following collections: AMNH; Australian Museum, Sydney; Auckland Institute and Museum, New Zealand; California Academy of Sciences; Emery coll.; Forel coll.; National Museum of Victoria, Australia; Paris Museum; Santschi coll.; ; Yasumatsu coll.

Apart from the holotype, all the specimens listed here have been designated as paratypes. The collectors G. Ettershank, T. E. Woodward and R. W. Taylor are referred to by their initials. Accession numbers following Ettershank or Taylor records refer to material in the MCZ collection. SAMOA: Upolu: Afiamalu, 700 m, (Type locality) beating, 1 dealate queen, 30.V1.1940 (E. C. Zimmerman); strays ex soil under stone, disturbed rain forest, 2 workers, 10.III.1962 (RWT acc. 245); nests ex rotting wood fragments (less than 6 cm diameter), rain forest floor, 10.III.1962 [RWT accs. 253 (holotype nest series — 10 workers, 6 males); 254 (9 workers); 259 (3 workers)]; nest under moss on rotting log, disturbed forest, 1 worker, 30.III.1962 (RWT acc. 632); ground moss berlesate, rain forest, 3 workers and 1 alate queen, 6.I.1956 (TEW); leafmold berlesate, rain forest, numerous workers, 16.III.1962 (RWT acc. 584); berlesates, moss ex ground and rotting logs, disturbed rain forest, III-IV.1962 (RWT accs. 302, 584, 2315-2319, 2322-2324); tree moss berlesates, disturbed rain forest, 19.I.1956 (TEW); III-IV.1962 (RWT accs. 300 < 2 m above ground); 582 (15 m); 580, 581, 2307, 2308 (all at 10 m above ground); GE accs. 37 (at 11 m); 44 (8 m above ground). Malololelei, ca 650 m, ex moss berlesate, rain forest, dealate queen, 19.I.1956 (TEW).

Afiamalu, 700 m, Upolu, Western Samoa, 10.III.1962 (R. W. Taylor). Holotype selected from a series of individuals taken nesting in a small fragment of rotting wood on floor of disturbed rain forest.