Crematogaster graeca, described new in 2026 — the mainland species that won the type table’s boldest row with another name’s excluded syntype, and whose oldest record was collected 114 years before its own holotype
R1010 closed on the other new species, the one that won without ever being measured as a type itself. This page reads that winner as the subject. Everything below is summarised in my own words, with no quoted text beyond bare collecting and bibliographic data.
First, the win that was not its own body’s. Table 2 of the revision runs six old type workers as wild cards against four clusters built from 201 workers in 68 nests. The highest score in the whole table, printed bold, is 0.990 — and it belongs to no type of graeca at all. It belongs to the Patras worker CASENT0908491, collected in April–May 1910, a syntype of Crematogaster ionia since 1911, measured digitally from its AntWeb photographs and assigned at 0.990 to the Greek-mainland cluster, against 0.001 to ionia itself. The ionia account excludes it from that name’s type series on that result; the graeca account then lists it first among graeca’s eight localities, still described there as a syntype worker of C. ionia. Graeca’s own holotype was never a wild card — Table 2 tests only old types — so the new species’ single most certain number is inherited: a verdict about a body that anchored another name for 115 years.
Second, what it is on its own. Crematogaster graeca was described as new in the same 2026 revision (ZooBank registration 932CE7B8-CB76-483B-8C8A-6B6FB02F48DC). Its holotype is a worker collected on 25 March 2024 by H. C. Wagner west of Piraeus, near Athens (37.9658° N, 23.558° E, 16 m above sea level), on a pine in a park, held at the NHMW in Vienna under the codename WAG3254 as the upper specimen of its pin; eleven paratypes from the holotype nest are spread across four collections. Behind the name stand 40 measured workers from 13 nests — the smallest sample of the four species, against 66 workers each for ariadnae and schmidti and 29 for ionia. It is the largest animal in the complex on the reported means: absolute cephalic size mean 1034 µm (range 941–1160), against 993 for ionia, 974 for schmidti and 971 for ariadnae. Its colour, in paraphrase, is a head, mesosoma and waist in homogeneous dark brown to reddish, with a dark brown to blackish gaster — a pattern it shares, in the key’s own wording, with both neighbours it must be told from.
Third, where it lives, and how deep its past runs. The account finds it at eight localities in Greece and North Macedonia, and nowhere else: Patras in 1910; the centre of Star Dojran and a site south of it in April 2010; and five 2024 sites — east-northeast of Astros, Arktisa, two sites north of Palea Epidauros, and the Piraeus park of the holotype — spanning, in the account’s regional words, Central Greece, Attica, the Peloponnese and North Macedonia. No locality in the study held two members of the complex together. The shape of the list is the point: seven of its eight records date to 2010 or 2024, the years of modern collecting, and the eighth predates them by a century. A species named in 2026 has a 1910 first record not because anyone saw it early, but because a body filed under another name was re-read. Its depth is real — the Patras worker was there, on that ground, in 1910 — and it is entirely retrospective.
Fourth, how hard it is to tell. The key reaches graeca last among the mainland pair, at couplet 3, for the Balkans south of 42° N: against schmidti it is told by a narrower petiole and postpetiole, a smaller postocular distance and a longer scape, with the couplet’s discriminant misplacing 5.0% of its 40 workers and none of its 13 nest means. Against ionia, one couplet earlier, it is told by wider frontal-carina and spine-tip distances, a longer mesosoma and shorter gastral pubescence (PLG/CS mean 0.063, against ionia’s complex-highest 0.070 and ariadnae’s 0.054). That ionia boundary is the revision’s tenderest: the pair’s cross-validation error, 5.8%, is the highest of the six pairs, against 1.9% for graeca–ariadnae and 4.7% for graeca–schmidti. In the full PCA its samples overlap a little with each of the other three; read in pairs, the plots clear — graeca against ionia alone separates cleanly, and with ionia removed it separates cleanly from ariadnae and schmidti together. The discussion leaves its eastern edge explicitly open: the transition zone with ionia in the strict sense, in Bulgaria, Turkish Thrace and northwest Anatolia, is named as future work.
Fifth, the life and the name. The account’s ecology is brief: nests on Eucalyptus, Pinus, olive and Cupressus, and once in a ruin wall; its trails were followed by three camponotines, Camponotus gestroi, Ca. kiesenwetteri and Ca. lateralis. The epithet, in paraphrase, refers to Greece itself, where most of the study’s investigated material came from and where the species is called common — a plain country name for the southernmost Balkan-mainland member, whose speciation, with its two fellow southern species, the discussion attributes to long isolation across the deep waters of the Aegean.
So what does the winner hold? Not a score of its own — its boldest number was earned by a borrowed body — and not a past of its own, since its first century exists only inside another name’s type series. What it holds is a transfer, completed: a cluster of 40 fresh workers coherent enough that a 1910 specimen could fall into it at 0.990, a holotype of its own from a Piraeus park, and a mainland range whose oldest point was waiting, mislabelled as someone else’s anchor, for 114 years. Ariadnae held everything fresh; graeca holds freshness and inheritance at once, and the inheritance is the older half.
The walk — five moments in one species’ double life, schematic only. Click each stage.
Pick a stage.
The runner-up needed no one’s type; the winner’s deepest record is one — certainty, here, was not made but handed over.