The one measured body in the schmidti file that anchors no name of its own — a Dalmatian worker from Ragusa, given by Mayr and held by him to be absolutely identical to a type too damaged to measure, which scored 0.945 in the type’s place
R1007 followed the synonym kept down by surplus and ended on the body that surplus quietly depends on. This page follows that body alone. Everything below is summarised in my own words, with no quoted text beyond bare collecting and bibliographic data.
First, the gap it fills. The name Crematogaster schmidti is anchored by Mayr’s 1853 type worker from the Vipava Valley, held in Vienna at the NHMW under the AntWeb identifier CASENT0919655. In the 2026 revision that type could not be measured at all: in paraphrase, it was not available for morphometric investigation because of damage, and it is held to its species by colour and zoogeography, which are said to match strongly. So the central name of the whole account has, in its own file, no number. Every number the account gives for schmidti’s types comes from some other body.
Second, what the stand-in is. The account introduces it, in paraphrase, as a further worker from Dalmatia. Its labels, paraphrased here, record that Mayr gave it, that according to him it is absolutely identical to the type, and that it is a cotype; they place it at Ragusa — Dubrovnik — in Dalmatia, collected by Reitter, and put it in Forel’s collection. It is housed in Geneva, at the MHNG, and its AntWeb pictures carry the identifier CASENT0908487. It anchors no name: it is not the schmidti type, not a synonym’s type, not the bearer of any epithet in the account. Its place in the file is entirely relational — it matters because of the body it is said to match. Dubrovnik also stands in the account’s list of eleven sampled schmidti localities, at 42.65° N, 18.09° E, credited to Reitter and to this same identifier.
Third, how it was judged. Like the medispina worker and the Patras worker, it was not measured on a stage in this revision; it was measured digitally from its AntWeb photographs and run as a wild card against the four clusters formed from 201 workers in 68 nests. It came back at 0.945 to schmidti. The rivals, from Table 2, are 0.031 to the Cretan ariadnae, 0.001 to the mainland graeca and 0.023 to ionia — 0.055 left standing for every other reading combined. Of the six type workers in the table, it is one of only three above 0.9. And here is the file’s small, unremarked split: the account’s text calls this worker a cotype, while Table 2 lists the same identifier as a schmidti paratype. Two different standing-words for one body, in the same revision, never reconciled on the page.
So does 0.945 confirm Mayr’s assertion? No — it agrees with it, and the difference is the whole page. The wild-card run does not compare this worker with the Vienna type; it cannot, because the type supplied no measurements to compare against. The run compares the worker with clusters built from 201 other workers, and returns: this body belongs with that crowd. Mayr’s label makes a different and stronger claim — not membership in a crowd, but identity with one particular damaged body in Vienna. That claim the number never touches. The chain that holds the name up has two separate links, fastened by different kinds of evidence: the type is tied to the cluster by colour and place, argued strongly and measured never; the stand-in is tied to the cluster by measurement, at 0.945, and tied to the type only by a dead man’s sentence on a label. Read strictly, the stand-in does not stand in for the type in the measurement — it stands in for it in the telling. Remove Mayr’s sentence and 0.945 is still true, and means less: a Dalmatian worker that fits the species. Keep the sentence and the number looks like a verification of the type itself, which it is not. What a stand-in holds, then, is a loan of authority in both directions — the type lends it relevance, it lends the type a number, and neither loan is ever called in, because the body that could settle the debt is the one body in the file that cannot be measured anymore.
The walk — five moments in one stand-in’s working life, schematic only. Click each stage.
Pick a stage.
The damaged type holds the name and cannot speak in numbers. The stand-in speaks at 0.945 and holds nothing but a sentence saying it is the same as the silent one — and the file stands, because the two of them together have exactly what neither has alone: an anchor, and a voice for it.