Tapinoma minor Bernard 1945

Seifert, Bernhard, Kaufmann, Bernard & Fraysse, Lorenzo, 2024, A taxonomic revision of the Palaearctic species of the ant genus Tapinoma Mayr 1861 (Hymenoptera: Formicidae), Zootaxa 5435 (1), pp. 1-74 : 49-53

Tapinoma minor Bernard 1945 [evaluation of type image and zoogeography]

This taxon has been described from the forest of Mamora near Ain Assou (34.17°N, 6.19° W, 93 m).

The image of a type specimen labelled “Mamora 1t 10 daya VI F.Bernard det., 42” [Bernard’s handwriting], “ Tapinoma minor 1942. Type ” [Bernard’s handwriting] and “ANTWEB CASENT 0913749” was investigated with making attempts to correct for measurement reductions produced by inadequate viewing positions. The following data could be derived: CS 655 µm, CL/CW 1.223, PoOc/CL 0.417, ExOcc/CS 0 %, SL/CS 1.003, ExCly/CS 10.52 %, ExClyW/CS 5.77 %, ExClL/W 1.823, dAN/CS 0.292, MGr/CS 3.47 %, MW/CS 0.659, Fu2L/CS 14.68 %, IFu2 1.489. When plotting all these shape variables against CS, each character is fully within that part of the cloud of T. simrothi & T. insularis n. sp. workers formed by the smallest individuals. Yet, a decision as to which of these two species the T. minor type should belong is not possible due to the lower accuracy of image-derived morphometry and lower reliability of single-specimen indication. The synonymization with T. simrothi is stated here because the type locality of T. minor is within the geographic range of T. simrothi but 1300 km west of the western range border of T. insularis n. sp. and in a natural habitat far away from a sea harbor.

Material examined. Numeric phenotypical data were taken in 40 samples with 121 workers. They originated from Algeria (1 sample), France (1), Italy (11), Morocco (4), Portugal (6), Spain (16) and Tunisia (1). For details see supplementary information SI1, SI2 .

Geographic range. West Mediterranean species, natural range reaching over 2000 km from Morocco (30°N, 10°W) east to Tunisia (35.7°N, 10°E) and going north to 41°N in Spain and Sardinia. The potential for anthropogenous spreading is apparently much lower than in the supercolonial species of the T. nigerrimum group. Yet, the northernmost known record from Faillant / France (43.489°N, 6.355°E) found in a small urban green within a strongly paved area should have been founded by anthropogenous introduction. The altitudinal range of 40 findings is characterized by 251 ± 379 [1, 1555] m with the highest site found in the Moroccan Atlas Mountains .

Diagnosis:—Worker ( Tab. 5): All shape ratios given below are, in contrast to those in Tab. 5, primary ratios without RAV and all data are given as arithmetic mean ± standard deviation. Medium-sized, CS 870 ± 94 µm. Head moderately elongated, CL/CW 1.094 ± 0.043. Postocular distance and excavation of hind margin of vertex rather large, PoOc/CL 0.395 ± 0.010, ExOcc 1.50 ± 0.82%. Anteromedian clypeal excision very deep and narrow, ExCly/ CS 12.74 ± 0.93%, ExClyW 6.01 ± 0.82%. The edge of clypeal excision curves slightly down below the level of adjacent clypeal surface, thus appearing rather blunt. Sum of pubescence hairs and smaller setae protruding across the margin of clypeal excision including its dorsal edge moderately large, nExCly 7.0 ± 3.5. Scape longer than in T. insularis n. sp., SL/CS 0.961 ± 0.026. Minimum distance of the inner margins of antennal socket rings rather small, dAN/CS 0.288 ± 0.006. Eye moderately long, EL/CS 0.253 ± 0.011. Metanotal groove moderately deep, MGr/CS 3.05 ± 0.68%. Mesosoma rather long and moderately wide, ML/CS 1.278 ± 0.032, MW/CS 0.636 ± 0.019. Second funiculus segment rather short, Fu2L/CS 13.46 ± 0.44%, IFu2 1.687 ± 0.106. All body parts including appendages covered by a moderately dense, short pubescence. Body surfaces more shiny than in the T. nigerrimum group. Setae on dorsal and lateral surfaces of head and mesosoma and on first two gaster tergites absent. Long seta are found on hind margin of 3rd and 4th gaster tergite, all gaster sternites, ventral parts of coxae and anterior clypeus. All body parts blackish brown. Mandibles and edge of clypeus sometimes with an orange or reddish color component.

—Male ( Fig. 7 ): the separation from the other species of the simrothi group by simple eye inspection of the genital appears difficult. Yet, a morphometric study appears promising when enough data will be available.

Taxonomic comments. The Tapinoma simrothi group contains four cryptic species separable by NUMOBAT: T. simrothi , T. insularis n. sp., T. phoenicaeum and T. karavaievi . Sufficient microsatellite data were available for the first three species and these genetic data confirmed the NUMOBAT classifications. Within a total of 43 samples for which both NUMOBAT and microsatellite data were available, the agreement of phenetic and genetic classification was 100%. Below we describe the run of exploratory and hypothesis-driven data analyses.

Block 1: Species hypothesis formation based on NUMOBAT data

The morphological separation of the four entities was done in several steps. In the first step of analysis all 16 NUMOBAT characters (CS and 15 RAV-corrected characters) were run unselectively in NC-part.kmeans and NCWard ( Fig. 47 ). The analysis of 132 samples revealed two big branches—one composed of clusters with West and Central Mediterranean distribution (later identified as T. simrothi and T. insularis n. sp.) and another one collecting clusters with more eastern distribution (later identified as T. phoenicaeum and T. karavaievi ). The classification of both exploratory data analyses for K=2 disagreed in 5.3% of the 132 samples. The final hypothesis for K=2 was determined by a controlling LDA in which the disagreeing samples were run as wild-cards. The error for K=2 was according to the controlling LDA 2.3% in NC-part.kmeans and 3.8% in NC-Ward.

As next step of analysis, the western and eastern cluster were analyzed separately according to the same schedule as reported above. The western cluster with 66 samples was separated into the species Tapinoma insulari s n. sp. and T. simrothi ( Fig.48 ). The classification proposals by NC-part.kmeans and NC-Ward disagreed with the controlling LDA in 1.5% and 7.6% of the samples. When run as wild-cards in a LDA, the neotype specimen of T. simrothi is allocated to the T. simrothi cluster with p=1.000 and the holotype sample of T. insularis n. sp. with p=0.9992 to the corresponding cluster.

The nest samples of the eastern cluster were separated into the species Tapinoma phoenicaeum and T. karavajevi ( Fig.49 ). The classification proposals by NC-part.kmeans and NC-Ward disagreed with the controlling LDA in 0% and 1.5% of the samples respectively. When run as wild-cards in a LDA, the lectotype sample of T. phoenicaeum is allocated to its name-bearing cluster with p=0.9992 and the lectotype sample of T. karavajevi with p=0.9998 to the corresponding cluster.

Block 2: Confirmation of morphological species hypotheses by nuDNA data

In Tapinoma simrothi , T. insularis n. sp. and T. phoenicaeum , where sufficient microsatellite data were available, 100% of NUMOBAT classifications could be confirmed by nuclear DNA. This was established by running a LDA using the first four principal components extracted from microsatellite data as characters and imposing the NUMOBAT classifications as hypotheses ( Fig.50 ). An overfitting of the LDA was not given as the smallest class contained 13 samples.

Biology. Apart from semi shaded conditions with trees such as olive plantations or cork-oak stands, T. simrothi prefers very sunny open habitats. These may be natural or semi natural (sand dunes, coastal cliffs, open patches in garrigue), rural (meadows, pastures, gardens, roadsides) or urban (areas with much pavement, asphalt or concrete and some greenery). Tapinoma simrothi is polygynous and may form both monodomous and polydomous colonies but apparently not such big supercolonies as known from species of the T. nigerrimum group. Nests are largely subterranean. Accurate and rather solid mounds of soil particles measuring up to 20 cm in diameter and height are typical above-ground structures in natural, semi natural and rural habitats but nests may be also under stones or in and under walls. Six observations of alates were made between 16 April and 30 May.

Figures

FIGURE 7. Male genital of Tapinoma simrothi in ventral view, Jamilena / Spain, 2014.04.23.

FIGURE 47. Demonstration of two main clusters within worker nest samples of the Tapinoma simrothi group by NC-Ward and NC-part.kmeans clustering considering 16 RAV-corrected phenotypical characters. Black cluster—T. insularis n.sp. et T. simrothi; gray cluster—T. phoenicaeum et T. karavajevi. The classification error of NC-part.kmeans relative to the controlling linear discriminant function is 2.3% for K=2.

FIGURE 48. Separation of worker nest samples of Tapinoma simrothi (grey bars) and T. insularis n.sp. (black bars) by NC- Ward and NC-part.kmeans clustering considering 16 RAV-corrected NUMOBAT characters. The classification error of NC-part. kmeans relative to the controlling linear discriminant function is 1.5%.

FIGURE 49. Separation of worker nest samples of Tapinoma karavajevi (red bars) and T. phoenicaeum (black bars) by NC- Ward and NC-part.kmeans clustering considering 16 RAV-corrected NUMOBAT characters. The classification error of NC-part. kmeans relative to the controlling linear discriminant function is 0%.

FIGURE 50. Comparison of the NUMOBAT-based classification of Tapinoma insularis n. sp. (black rhombs), T. simrothi (white triangles) and T. phoenicaeum (black dots) with the genetic classification based on 15 microsatellite loci. Method: the first four principal components extracted from microsatellite data of only the shown 43 samples were used as characters in a linear discriminant function with the NUMOBAT-based classification imposed as species hypothesis.