A taxonomic revision of the Palaearctic members of the Formica rufa group (Hymenoptera: Formicidae) - the famous mound-building red wood ants Author Seifert, Bernhard text Myrmecological News 2021 2021-04-28 31 133 179 http://zoobank.org/0e55c0d7-531a-48d7-a078-148b96bd461d journal article 10.25849/myrmecol.news_031:133 b9f36fb1-1c9d-4af8-96ca-d57973b94862 1997-3500 5582216 0E55C0D7-531A-48D7-A078-148B96BD461D Formica paralugubris SEIFERT, 1996 Formica paralugubris SEIFERT, 1996 [ type investigation] This taxon was described from the Swiss Jura Mountains ( 46.537° N , 6.192° E, 1450 m ). The holotype gyne is labelled “SWI: Jura : 1994.06, Le Brassus-5SSW, Chalet a Roch Field Stat., nest G5” and depicted in AntWeb ( ANTWEB 2021 ) under the unique specimen identifier FOCOL0762 . Investigated was all type material, consisting of five gynes and 34 workers from the nests G1-G5 of the holotype supercolony, collected in the years 1993 and 1994. Depository SMN Görlitz. All material examined. Numeric phenotypical data were recorded in 73 nest samples with 355 workers and in 53 gynes. These originated from Austria (six samples), Canada (two), France (two), Germany (four), Italy (three), and Switzerland (58). Geographical range. Its natural range is rather small and extends over the montane-subalpine zone of the Jura Mountains and western Alps between 6° E and 11.5° E with a small exclave in the southern Schwarzwald / Germany . In the Alps, it ascends to 2300 m . A colony artificially introduced to Quebec / Canada in 1971 showed continuous growth to supercolonial size over 34 years ( SEIFERT 2016a ). Artificial introductions of wood ants to at least 42 localities over entire Italy south to Sicily and west to Sardinia were performed in the years 1959 - 1967 (e.g., PAVAN 1959 ). In that time, the transferred ants were classified as Formica lugubris . However, it is very likely that the vast majority of these introductions really involved Formica paralugubris as it was confirmed for five sites in the North Apennine ( MASONI & al. 2019 ). Diagnosis of worker ( Tab. 4 , key). Minimum size, mean and maximum CS 1680 and 2020µ m.Head rather short, CL / CW 1750 1.091. Scape rather short and thickset, SL / CS 1750 0.902, SL / Smax 1750 9.22. Eyes always with long microsetae, EyeHL 1750 34µm . Setae number on dorsal plane of scape variable but on average higher than in Alpine Formica lugubris , nSc 1750 5.2. Posterior margin and underside of head always with conspicuous setae, nCH 1750 24.9, OccHL 1750 108 µm , nGu 1750 14.2, GuHL 1750 164µm . Mean length of pronotal setae, number and length of metapleural setae on average lower than in morph A1 of Alpine F. lugubris , mPnHL 1750 78 µm , nMet 1750 7.7, MetHL 1750 154 µm . Workers of morph A3 of Alpine F. lugubris are similar in the pilosity condition but have a much larger size, a larger head length index, and a shorter scape. Diagnosis of gyne ( Tab. 7 ). On average smaller than morph A1 and A3 of Alpine Formica lugubris , mean and maximum CS 2095 and 2238 µm . Scape longer than in morph A3 of Alpine F. lugubris and very thickset, SL / CS 0.805, SL / Smax 7.97. Eyes always with conspicuous microsetae, EyeHL 41 µm . Setae number on dorsal plane of scape variable but on average higher than in morph A1 and A3 of Alpine F. lugubris , nSc 6.4. Posterior margin and underside of head always with conspicuous setae, the length of which is lower than in morph A1 but larger than in morph A3 of Alpine F. lugubris , nCH 23.8, OccHL 117 µm , nGu 16.7, GuHL 128 µm . Pronotal setae shorter than in morph A1 of Alpine F. lugubris , mPnHL 88 µm . Petiole setae fewer and metapleural setae shorter than in morph A1 of Alpine F. lugubris but more numerous and longer than in morph A3 of Alpine F. lugubris , nPe 9.2, MetHL 110 µm . Dorsal surface of gaster appears at lower magnification more or less shiny. Dorsum of first gaster tergite usually with weak transverse microripples and with foveolae and pubescence on average more densely packed than in Formica rufa or Formica polyctena , FodG 21.1 µm , sqPDG 4.62 µm . Taxonomic comments and clustering results. Considering the extreme polymorphism in Alpine Formica lugubris and the presence of another similar sympatric species Formica helvetica sp.n. , the separation of Formica paralugubris in both workers and gynes should be problematic. I combined 98 nest samples with 409 workers of Alpine F. lugubris morphs A1 and A3 and of F. helvetica sp.n. in class 1, and 70 nest samples with 323 workers of F. paralugubris in class 2. A two-class LDA considering the characters CS , CL / CW 1750 , SL / CS 1750 , SL / Smax 1750 , PeW / CS 1750 , nSc 1750 , nCH 1750 , OccHL 1750 , mPnHL 1750 , nMet 1750 , and MetHL 1750 classified all samples of F. paralugubris and 96 samples of the collective cluster correctly. This means a classification error of 1.2% within a total of 168 nest samples. A plot of the first and second factors of a PCA supported the existence of two main clusters class 1 and class 2, and disagreed in 3.0% of the samples with the LDA ( Fig.32 ). The exploratory data analyses NC-part.kmeans, NC-Ward, and NC-NMDS.kmeans suggested two clusters and disagreed with the final species hy- pothesis by 3.6, 4.1, and 3.0%. The clustering by NC-part. hclust was ignored as it splitted into seven clusters and exposed 8.9% indeterminate samples (outliers). As result, we have a sufficiently good separation of F. paralugubris workers by morphology. The distinction of F.paralugubris gynes from those of Alpine F.lugubris morphs A1 and A3, and of F. helvetica sp.n. by a principal component analysis appears also rather clear when the eight characters CS , SL / CS , SL / Smax, PeW / CS , ML / CS , nSc, nCH, and OccHL are considered ( Fig.33 ). Section “Hybrids Formica aquilonia × paralugubris ” (p.175) discusses the situation in hybrids F. aquilonia × paralugubris . Fig.32: Nest-sample means of the first and second principal component of workers of Formica paralugubris (black dots), Formica lugubris morph A1 (squares), F. lugubris morph A3 (triangles), and Formica helvetica sp.n. (rhombs). Eleven phenotypic characters were considered. Biology. See the species profile given by SEIFERT (2018) .