The chameleon

Colobopsis imitans read whole — the other follower, a second ant that converged on the same trails and the same models from a different genus, hidden for two centuries inside another species’ name and described new only in 2022

R1015 closed its page by pointing at this animal and asking what two followers hold when neither invented the following. Camponotus lateralis, the follower of the previous page, has been following Crematogaster trails in print since 1842. This one followed the same trails, in the same colours, on the same trees — and was not even a species on paper until 2022, because everyone who collected it filed it under its sister’s name. Read whole, it is the strongest evidence in the sequence that the models’ colours are a shelter worth evolving twice — and the one case where the copying became so versatile that it nearly dissolved the copier’s own identity. Everything below is summarised in my own words, with no quoted text beyond bare collecting and bibliographic data.

First, the hiding. For a long time every West Palaearctic Colobopsis was classified as a single species, Colobopsis truncata (Spinola, 1808) — an arboreal formicine whose phragmotic major workers block the nest entrance with their flattened heads, and which is itself a Batesian mimic, of the dolichoderine Dolichoderus quadripunctatus, whose trails it follows only rarely. Scattered accounts of Colobopsis ants mimicking other models existed but were never addressed comprehensively. Schifani and colleagues (2022, in the Zoological Journal of the Linnean Society) tested whether more than one species was hiding in the region, combining morphology, colour, genetics, behaviour and distribution. The answer was a second species, and they named it for the trick itself: imitans, the present participle of the Latin verb meaning to imitate. The holotype worker and fourteen paratypes came from Mondello, Sicily (38.1953, 13.3354, 5 m, collected 14.X.2018), and the holotype is stored in the Hungarian Natural History Museum. The new species was set apart from truncata by a constant colour pattern very close to Crematogaster scutellaris, by cryptic but reliable morphometric differences, by genetics, and by behaviour: where truncata does not regularly follow trails, imitans follows scutellaris trails extremely frequently and coexists with its model on the same trees. Its range as first drawn was the southwestern Mediterranean — Sicily, southern Iberia and the Maghreb — matching, as the describers noted, the distribution of the ant it mimics.

Second, the widening. That tidy picture lasted three years. Wagner & Csősz (2025), working from morphometric samples across the Mediterranean Basin and the Caucasus and documenting the colour patterns of mimics against their local models, found imitans replacing truncata throughout southern Europe — and wearing not one uniform but several. Alongside the scutellaris-like form of the original description there is a Dolichoderus quadripunctatus-like morph, in central Iberia, southern Greece, Crimea and the Caucasus east to Azerbaijan, which mimics the very species truncata mimics and is chromatically very similar to truncata itself; and an ionia-like, blackish morph in Anatolia, Crete and Cyprus, mimicking Crematogaster ionia and following its trails. Their central finding is the one that earns the animal its newer name: unsupervised multivariate analyses found no morphometric discontinuities separating these regional colour forms into lineages. The diversity is mimicry-driven microevolution — significant phenotypic variety produced by adjusting to locally available models, without speciation. Hence the title of their paper calls it a true chameleon, in a biogeographical sense.

Third, the edge of the map. In August 2025 a single worker was collected from a Crematogaster ionia trail on a Lebanon cedar oak (Quercus libani) near Saghbine, in the Beqaa region of Lebanon (33.61305, 35.68862, 23 Aug 2025) — the first record of the species in the Levant and the southernmost record anywhere, 300–365 km from the nearest confirmed records in Cyprus and Türkiye, while the closest confirmed truncata are some 1200–1425 km away in Greece (Schifani & Massaad 2026). The worker belonged to the dark, ionia-mimicking morph: dark brown across the body and appendages except for a black gaster, no white stripes on the gaster, closely matching the ionia workers taken from the same tree — though lighter, browner, than the blackish morph as photographed from Crete, much as the Anatolian and Samos material is also browner than properly blackish. Its identity was settled not by its colours — colours are exactly what this species cannot be trusted on — but by measurement: seven morphometric characters fed into the DColo discriminant function of Wagner & Csősz gave a score of 1.485, safely on the imitans side of zero and outside the overlap range (the function’s stated error rate is 5.6% in individual workers). The same paper draws the honest consequence: because an imitans morph now exists that looks like truncata, the old truncata records from Jordan, Lebanon and Palestine are doubtful, and many published under the one-morph assumption likely belong to imitans. The chameleon does not just wear other ants’ colours. Retroactively, it wears its sister’s records too.

Fourth, the two followers compared — and the comparison that does not exist. R1015 asked which of the two followers is the better copy where their ranges overlap. The honest answer is that the literature has never scored them the same way. Camponotus lateralis has been measured against its models: 957 mimics allocated as wild cards to the three Crematogaster colour clusters, correct at 66.0%, 98.5% and 88.0% beside its three models. Colobopsis imitans has been measured against its sister: separated from truncata by a discriminant function with error rates of a few percent, and by the unsupervised finding that its own colour forms do not split it. One follower is graded on resemblance to what it copies; the other is graded on distinguishability from what it is not. What can be said is structural. Lateralis keeps one name across three regional uniforms and was never hidden; imitans wore its uniform so well, for so long, that it had no name of its own until 2022 — and then, within three years, its one uniform became at least three, one of them borrowed from its sister’s model, so that the species defined by imitating scutellaris turns out also to imitate the ant its sister imitates. Convergence is the point R1015 wanted tested: two formicine genera, independently, arrived at the same trails, the same models and the same regional wardrobe. The models’ colours are, on this evidence, a shelter two separate lineages paid evolution’s price to enter.

A five-stage walk (schematic only — the stages are this page’s arrangement, not a scheme from any paper):

hidden inside its sister’s name, to 2022
described for the imitation itself, Sicily, 2022
one model, one uniform: scutellaris in the southwest
three uniforms, no new species, 2025
a dark worker in Lebanon, measured not seen, 2025

Pick a stage.

The models hold a coast, an island, a narrowed name and the position of being copied. The first follower holds the copying at sixty-six, eighty-eight and ninety-eight point five percent. The second holds it so completely that for two hundred years nobody could see there was a second animal there at all — and what it holds now, under its own name at last, is a wardrobe: whichever model the coast provides, worn without becoming a new species for any of them.