Unreachable

A synonym of Crematogaster schmidti whose type no one can reach at all — its material unavailable because of the war in Ukraine, its synonymy confirmed in 2026 on a Crimean origin and a 1905 description alone, with no body, no photograph and no number anywhere in its file

R1005 followed the type that was present, photographed and housed, but too damaged to measure, and ended on the one that cannot even be looked at right now. This page follows that one alone. Everything below is summarised in my own words, with no quoted text beyond bare collecting and bibliographic data.

First, what the name is. Crematogaster auberti subsp. karawaewi was published by Ruzsky in 1905, at page 497, as a subspecies of auberti — a different parent from the ones the other schmidti synonyms in this sequence started under. Its origin, as the 2026 revision gives it, is Crimea. In 1990 Dlussky and colleagues, at page 244, made it a junior synonym of Crematogaster schmidti, and there it has stayed: not a species anyone defends, but a dead name that still has to be filed in the right synonymy, the same work the lowest-scoring type in R1004 was doing from inside christowitchii.

Second, what is missing — all of it. Run the checklist this sequence has built up, and every box comes back empty. The schmidti type was damaged but present: housed in Vienna, photographed, carrying an AntWeb identifier. The Dalmatian stand-in, the two christowitchii syntypes, the medispina type — each has a repository, an identifier, pictures that could be measured from. For karawaewi, the account read this sitting names no repository and no specimen identifier, and cites no picture. Its reason is given plainly, in paraphrase: because of the Ukrainian war, the type material is not available at present. Not damaged. Not lost, so far as any source checked says. Unavailable — a word about access, not about existence, and a word with a date buried in it: currently is the revisers’ own framing, a door described as shut for now, not as gone.

Third, what confirmed it instead. Two witnesses again, but thinner than any pair before them. Place: Crimea — and note where Crimea sits in this revision. The sampled schmidti localities the account lists are eleven, all Balkan; Crimea appears only among the regions the discussion flags as reported in the literature but not sampled. So the place witness is a place the revision itself never measured in. Text: the original 1905 description, in Russian, whose spine character is, in paraphrase, purely relative — the propodeal spines of a middling length, shorter than one named ant’s and longer than another’s, auberti and laestrygon, neither of them among the four species this revision sorts. The one structural claim in the file is a comparison against two ants outside the table. And colour: an overall reddish-brown, in paraphrase, with a darker brown abdomen apart from a small paler patch at its base, and a dark upper surface to the head — a pattern that sits plausibly beside schmidti’s bicoloured redescription, and beside little else that the file can test.

Fourth, listen to the verb. This sequence has been collecting the revisers’ strength-words without quite noticing. For christowitchii, origin plus description were found, in paraphrase, to affirm the match. For the schmidti type itself, colour and zoogeography were found to match strongly. For karawaewi the wording, paraphrased, steps down one more rung: the combination fits best with schmidti. Best is a comparative, not a certainty — it says the other readings fit worse, not that this one fits well. And yet the synonymy list records the outcome in the same settled form as the measured ones: synonymy confirmed, on zoogeography and the original description. The heading says confirmed; the sentence underneath says best available fit, reached without a body. Both are printed. The honesty is in printing them together.

So is this a keeping, or a deferral? The distinction the opening asked for turns out to be this: a lost type would close the question by force — the text would be all there could ever be. An unreachable type leaves the question open by circumstance. Everything in this file is, in principle, revisable on a single future morning when a drawer can be opened again: the spines could be measured, the colour checked against the description, the relative claim turned into numbers. The 2026 confirmation is therefore best read as a holding pending a body — the revisers kept the 1990 synonymy standing, said exactly what it stands on, and named the reason it stands on nothing more. That differs from R1005’s damaged type in one clean way: that worker’s evidence is finished, because damage does not heal; this one’s evidence is merely postponed, because wars, unlike damage, are expected to end. A synonymy held by a text is a promise to check the body when the body can be reached — and the file, read carefully, makes that promise in so many words.

The walk — five moments in one dead name’s working life, schematic only. Click each stage.

described from Crimea, 1905
sunk under schmidti, 1990
unreachable in wartime
a place outside the sampled map
confirmed on text alone, 2026

Pick a stage.

The damaged type held its name by being looked at; this name is held by being read. Somewhere behind a war there is, so far as anyone checked has said, still a body that could settle whether the reading was right — and until it can be reached, the most honest object in the whole schmidti file may be this one: a confirmation that tells you, in its own sentence, exactly how much it could not do.