Ants of Tawau and Borneo — Formicidae

Records of ants photographed in and around Tawau, Sabah, by Wong Fook Yee. This page gathers them from across the Insects/ folders and says plainly what each record does and does not establish.

What makes an ant an ant

Ants are wasps that took up living in colonies. They are a single family, Formicidae, inside the order Hymenoptera, and about 13,800 species have been described. Three characters separate them from every other insect, and two of the three can often be read straight off a photograph:

Geniculate antennae. The antenna has a long first segment, the scape, and then bends at a sharp elbow into the rest. Nothing else in the insects carries antennae quite like it, and it is the single most useful thing to look for when you suspect you are being shown a mimic.
A node-like waist. The second abdominal segment is strongly constricted into a petiole bearing one or two distinct nodes. A real ant's waist is a genuine narrowing with a lump on it, not a pale band across an unbroken body.
The metapleural gland, which produces antibiotic secretions that keep a crowded nest free of disease. It is unique to ants among all insects — but it cannot be seen in a field photograph, and some genera, Camponotus among them, have lost it.

Those first two characters are why this site can be confident that an animal is an ant while still being unable to say which ant. They establish the family and go no further.

Why this index sits above the wing-count folders

The Insects/ section of this site is filed by wing count — 0_Wing/, 1_pairWing/, 2_pairWings/. That key works well and is kept. Ants break it, and they break it in a way worth understanding, because for an ant the number of wings is not a species character at all. It is a caste character.

Workers are always wingless, in every ant species without exception. They file under 0_Wing/. Nearly every ant anyone photographs is a worker.
Queens and males have two pairs of wings. They fly once, on the nuptial flight, which in Borneo often follows heavy rain and brings them to lighted windows in numbers. A winged ant belongs under 2_pairWings/.
A mated queen is wingless again. She sheds her wings after the flight, leaving visible stubs on a deep, boxy thorax built for flight muscle. She returns to 0_Wing/ — the same individual, filed in two different folders at two points in her life.

So one species can generate records in two folders, and one animal can too. Rather than split the family, this index sits above the wing-count folders and reaches down into all of them — the same arrangement already used for the true bugs and the stick insects.

The records

For students — a page written to teach

The green queen and the maggots — a weaver ant on a papaya leaf
19 July 2009, 17:31. A papaya garden beside the Tawau River opposite Pasar Semarak. A dead queen weaver ant, about 15 mm and green, lying on a sheet of silk across a dying papaya leaf with pale fly larvae on her. Most probably Oecophylla smaragdina — the kerengga, whose workers are red but whose queens are green.
This page is written for Malaysian students rather than for specialists. Its photographs are labelled and reduced, and its subject is as much how to think as what to name: how to tell a queen from a worker, a maggot from an ant larva, and how to say plainly that you do not know whether the larvae killed her or found her already dead. — iNaturalist observation 398990434

Formicinae — named to species

The Fish Hook Ant — the fig garden, Tawau Hills
4 June 2023. About 10 mm, orange-brown under a dense golden pile, with black legs, black antennae and black spine tips. Seen walking fast and alone in low bush in the fig garden opposite Tawau Hills Park; photographed the same night in a container, which is why the background is white.
Polyrhachis bihamata (Drury, 1773) — the first ant on this site carried to species, and it gets there because the deciding character is finally visible: a pair of long hook-shaped spines lying parallel on the waist. Bihamata means "two-hooked". The three look-alikes (P. dostali, P. lacroixi, P. tonsilis) are separated on strongly reduced standing hairs, which this animal does not have.
Worth noting: the Malaysia Biodiversity Information System lists this species from Negeri Sembilan and Perak only, with no Sabah locality. Written for visitors rather than for specialists. — iNaturalist observation 399103222

When a weaver ant met a nettle caterpillar — Kampung Sayap, Kota Belud
20 December 2022, 10:48. A blade of long grass at Eco-Edu Tourism Tinuman, on the northern flank of Mount Kinabalu — the first record on this site from outside the Tawau district. A major worker of Oecophylla smaragdina, the kerengga, walked up to a 10 mm nettle caterpillar (Lepidoptera: Limacodidae, genus undetermined), touched it with its antennae, and left it alone.
The subject of the page is the meeting, not either animal. The caterpillar's hollow, venom-filled spines hold an ant's jaws away from any surface it could grip, and the ant established that by touch without ever being stung — a defence that works by being recognised rather than used. The two frames are 56 seconds apart: ant present, then gone. Written for visitors. — iNaturalist observation 399882665

Formicinae — genus determined

A spiny ant on dead wood — the fig garden site, Tawau Hills
4 September 2015, 13:51. About 10 mm, black and strongly shining, walking alone on a split piece of rotten wood on the same strip opposite Tawau Hills Park — ninety-four minutes after the record below, and a few paces away. Recorded at genus level: Polyrhachis, the spiny ants, on the spined and sharply margined thorax. Echinopla and Camponotus are both excluded because neither carries thoracic spines. The species is left open. A second frame was taken from above specifically to show the waist, whose spines separate the species; it shows the waist as a low unarmed node but its corners are crossed by a leg and in shadow, so no count is claimed. Two small red bodies, probably mites, ride on the legs. — iNaturalist observation 398945387

Formicinae — probable

A black ant on a grass stem — the fig garden site, Tawau Hills
4 September 2015, 12:17. About 10 mm, black and shiny, with very long bristled legs, walking alone along a grass stem on the strip of ground opposite Tawau Hills Park. Almost certainly a worker. Recorded at family level; the waist appears to carry a single node, which would put it in Formicinae. Camponotus and Polyrhachis are the two candidate genera and the page explains why the photograph cannot choose between them — and the record above, taken on the same strip the same afternoon, is a Polyrhachis beyond doubt, which is worth comparing against. Its second value is as a dated record of that half-kilometre six years before it was planted as the Plant4Tawau fig garden. — iNaturalist observation 398063624

Not an ant, but filed here on purpose

An ant that is not an ant — Ikutmaju Saw Mill, Merotai
23 October 2014. A true bug, not an ant. Its narrow waist is a pair of pale markings rather than a real constriction, and the photographer's own field note recorded "two segments, the head and body", which is the mimicry succeeding on an experienced observer at close range. It is listed here because anyone using this index to name a black insect with a narrow waist needs to meet it. Ant mimicry has evolved independently more than seventy times.

The older "Ant of Sabah" page

An earlier page at 0_Wing/INDEXAnt.htm, last edited in January 2016, presented ten photographs of Sabah ants as a row of thumbnails with no names, no localities and no dates. That address now redirects here.

All ten photographs are missing from the server. Checked on 7 September 2026: Ant_01_small.jpg, Ant_01.jpg, Ant_05_small.jpg and Ant_10.jpg all return 404, against a control image on the same folder that loads normally. The page has been showing ten broken thumbnails to anyone who followed the "ANTS" link from the insects index.

The photographs themselves are very likely still in the photographer's own archive. If they are, each one can become a record here on the same terms as the record above — a locality, a date, a size, and an honest statement of how far the identification goes. Ten unnamed thumbnails are worth little; ten dated field records from Sabah are worth a great deal.

How much of this is still unnamed

The standing checklist of Bornean ants records 717 valid species and 52 further subspecies, in 97 genera and 12 subfamilies, and its authors estimate that at least 1,100 species are actually present on the island. Two subfamilies hold most of them: Myrmicinae with 315 species and Formicinae with 213. The largest single genus is Polyrhachis, the spiny ants, with 98 recorded Bornean species; Pheidole, Camponotus, Crematogaster and Strumigenys are also well represented, and half of all Bornean ant genera contain only one or two species each.

This is the background against which the caution on these pages should be read. A photograph of a black ant on a leaf, taken with a compact camera, is not evidence for a species name when the island holds a hundred species in one genus alone. Family is often as far as it goes, and family, honestly stated and attached to a date and a place, is a real record.

References used across these pages

Ant — family-level summary: geniculate antennae, the metapleural gland and its loss in Camponotus, the node-like petiole, wingless workers, winged reproductives, queens shedding their wings after the nuptial flight, and the approximate world species total quoted above.
Polyrhachis, AntWiki — the pronotal, propodeal and petiolar armature that separates the spiny ants from Camponotus.
Pfeiffer, M., Mezger, D., Hosoishi, S., Yahya, B.E. & Kohout, R.J. (2011), "The Formicidae of Borneo (Insecta: Hymenoptera): a preliminary species list", Asian Myrmecology 4(1): 9–58 — the Bornean species, genus and subfamily totals quoted above.
iNaturalist — where these observations are open to identification by specialists. Observer: wongfookyee.