The research was carried out at ten sites in Czechia (Table 1, Fig. 1). In one site, Suchomasty, we documented temperature fluctuations measured by dataloggers Tinytag Ultra2 every three hours at the maximum sampling depth and on the surface (Fig. 5).

Figure 1. 

Locations of the study sites in Czechia. 1 – Kostelní Bříza, 2 – Raná, 3 – Suchomasty, 4 – Nebušice, 5 – Ctiněves, 6 – Hostovice, 7 – Mravín, 8 – Pouzdřany, 9 – Dolní Věstonice, 10 – Kurovice. For details, see Table 1.

The plastic pipes (for lengths see Table 1) had an inner diameter of 7 cm and were perforated by oblique, 5 mm wide cuts. Plastic cups were mounted onto a threaded metal rod at 10 cm distances (for example, a 2-m long pipe contained 20 cups) (Fig. 2A) and contained a solution of 6% formaldehyde with few drops of detergent (for more details see Růžička and Dolanský 2016). We installed these traps in excavated holes or boreholes (a drilling machine was used in Suchomasty) and emptied them once a year. When we installed multiple pipes per site, each pipe was buried in a separate hole. Since traps were buried for long periods (between 2 and 8 years depending on the site), we expect the initial mixing of the soil layers to exert a minor role in the distribution of fauna.

Figure 2. 

A a set of plastic cups taken from the pipe trap in the Carpinus forest on limestone near Suchomasty (photo M. Šafra); the soil profile is inhabited by Porrhomma cambridgei, see Fig. 3A B a fissure network in sandy marl near Nebušice, which was rich in Porrhomma microcavense, see Fig. 3D (the pipe trap was ten meters from this quarry wall) (photo M. Řezáč) C, D a scree soil near Hostovice, which was rich in Porrhomma microps, see Fig. 3E (photo J. Dolanský) E a fissure network in sandy marl near Mravín (the pipe trap was 200 meters from this profile) (photo J. Dolanský) F a hole for a pipe trap near Mravín in the soil, which was rich in Iberina microphthalma and P. cambridgei, see Fig. 3C (photo J. Dolanský).

Spiders were sorted and examined under an Olympus SZX12 stereomicroscope in 80% ethanol. Spider identification follows Nentwig et al. (2022), and nomenclature is in accordance with the World Spider Catalog (2022). The individuals were deposited at the institutions of the authors. Photographs were taken with an Olympus C-5060 wide-zoom digital camera mounted on an Olympus BX40 microscope. The images were processed using CombineZP image stacking software.

We classified the affinity of particular species to subterranean environments according to their vertical distribution. Further, we considered (1) morphological adaptations for subterranean life, (2) habitat, where they were found so far, (3) pattern of their distribution (common or rare on the surface habitats), (4) what environment they need for their lifestyle (see column “relation to the subterranean habitat – RSH” in Table 2). Morphological features, in particular depigmentation and a reduction in eye size, were recorded during identification, and the carapace size data were taken from the literature (Merrett et al. 1993; Weiss 1996; Růžička 2018, 2022; Šestáková et al. 2018; Nentwig et al. 2022). Data on spider occurrence in Czechia, habitat, and natural history were taken from a database of records of spiders in Czechia (Czech Arachnological Society 2022) and Buchar and Růžička (2002). The following categories were distinguished:

1. surface species occur only at the surface, on the ground, or in the vegetation. Their presence underground is accidental; they possess eyes of normal size and are normally pigmented.
2. detritus layer species live in detritus on the soil surface. They are rare in deeper soil layers. They possess eyes of normal size and are normally pigmented. Two special cases are within this category:

1. species with burrows in soil spend at least part of their lives in shallow burrows in soil; they possess eyes of normal size and are normally pigmented.
2. species requiring high air humidity require habitats with constantly high air humidity, regardless of whether the habitats are superficial or subterranean; they possess eyes of normal size and are normally pigmented.

1. soil species are more common deeper in humus soil than in detritus, but they appear on the soil surface relatively often; they are usually tiny and less pigmented, and their eyes are relatively smaller than those of their relatives. A special case is within this category:

1. soil ant nest species spend the majority of their lives in ant nests in the soil; they are less pigmented, and their eyes are relatively smaller than those of their surface relatives.

1. soil-bedrock species live in the basal mineral layers of soil on eroded bedrock surfaces (or in other spacious subterranean habitats) and appear on the soil surface very rarely; they are pale, and their eyes are significantly reduced; they can be larger with relatively long legs.

We used descriptive statistics [mean, standard error (SEM), and 95% confidence interval] to assess the species’ preference for surface versus subterranean habitats, approximated via the depth of capture of each individual. In cases where the whole confidence interval was below a depth of 20 cm (the depth of the two uppermost cups in the trap, where the surface species can easily enter), we assumed the species to be more common below 20 cm depth than above it (the species shown in bold in Table 2). Data were handled in R, version 3.6.2 (R Core Team 2019).

The captured subterranean species belonged to the families Linyphiidae, Hahniidae, Theridiidae, and Nesticidae, with Linyphiidae having the greatest abundance and species richness. Three genera dominate in the Central European shallow subterranean habitats: Centromerus, Palliduphantes, and especially Porrhomma. At two study sites, Mravín and Suchomasty, Porrhomma cambridgei and Porrhomma microps were found to occur syntopically. In both cases, the smaller P. cambridgei was found deeper than the larger P. microps (Fig. 3A, C). So far, Porrhomma campbelli has been known from the moss in wet habitats, such as peat bogs, spruce forests, and brook and pond margins (Růžička 2018). Here, we present the first finding of this species from subterranean habitats in Czechia.

The genus Centromerus was represented by four species, two of which, C. cf. piccolo and C. cf. minutissimus, were new to Czechia. We captured 17 specimens of Centromerus cf. piccolo at depths of 50–90 cm in loess accumulations in south Moravia. Previously, Centromerus piccolo had been known only from Germany (Weiss 1996; Nentwig et al. 2022). As the genitalic morphology of our specimens slightly differs from the illustrated type specimens of C. piccolo (Weiss 1996), we can not exclude the possibility that this is a closely related new species. In the artificial gravel accumulation in western Bohemia, we found two Centromerus females that are not described in the literature. Based on somatic characters (including small body size, eye reduction, and depigmentation) and the fact that the only Central European Centromerus species with unknown females was Centromerus minutissimus, which were known from England, Germany (the site in southern Saxony-Anhalt is only 200 km northwest of our site), south France and Spain (Merrett et al. 1993; Cárdenas and Barrientos 2011; Nentwig et al. 2022, GBIF – www.gbif.org, INPN – www.inpn.mnhn.fr), we preliminarily determined these individuals to be Centromerus cf. minutissimus.

We captured three species of the genus Palliduphantes. Palliduphantes insignis has been found in Czechia very rarely, only on the steppes on rocks or loess in warm areas (Buchar and Růžička 2002). We found 31 specimens at a depth of approximately 30 cm in loess. Thus, its preferred microhabitats are probably micro spaces in soil and fissures in bedrock. Palliduphantes pallidus and Palliduphantes alutacius sensu Miller and Obrtel (1975) are closely related taxa that are not morphologically clearly separated in Czechia. Instead, there are mainly typical Palliduphantes pallidus in the western part of Czechia, while in the eastern part typical Palliduphantes alutacius occurs, and morphologically intermediate forms can be found between. This pattern probably demonstrates the introgression of these two taxa in this territory. These species are among the most common subterranean species in Czechia and are also often found at the surface.

Concerning the remaining subterranean linyphiid spiders, we found four species of the subfamily Erigoninae: Jacksonella falconeri, Mioxena blanda, Pseudomaro aenigmaticus, and Tapinocyboides pygmaeus. Singletons of these species have rarely been found on the surface, usually in open xerothermic habitats (Czech Arachnological Society 2022). Our data show that the main microhabitats of these species are interspaces in various substrata.

The subterranean Hahniidae spiders were represented by three species. European records of Iberina microphthalma were summarized by Růžička and Dolanský (2016). Furthermore, in Czechia, this species was collected in the substratum on claystone, sandy marlstone, or sandy marl (Růžička 2022). We found 17 specimens in the soil on sandy marl at depths of 40–120 cm (Fig. 3C). Cicurina cicur is common under stones, in leaf litter, and decaying wood in forests, as well as in shallow subterranean habitats (Buchar and Růžička 2002; Růžička and Dolanský 2016). Because of its morphological appearance – depigmentation and small eyes – it is expected for this species to spend most of its life in the subterranean environment, especially in southern Europe (Mammola et al. 2022). Indeed, we found it to be dominant in the studied shallow subterranean habitats.

Nesticus cellulanus is the only representative of Nesticidae, the spiders mainly occurring in subterranean environments (Mammola et al. 2022). As this species is relatively large, it mainly occurs in caves, scree slopes, or man-made subterranean spaces, whereas it is rare in interspaces in the substratum.

The family Theridiidae was represented by Theonoe minutissima and some species of the genus Robertus. Theonoe minutissima regularly occurs in scree slopes, we found it only inside artificial gravel accumulation. Robertus species recorded during our study are common in detritus.

At two sites, Kostelní Bříza and Mravín, we captured the relatively large depigmented hahniid spider Mastigusa arietina, which is known to live in subterranean ant nests (Nentwig et al. 2022). At the Mravín site, it was found with another myrmecophilous spider, Syedra myrmicarum. The dominant co-occurring ant species was Lasius flavus (Fabricius, 1782).

The body size and relative length of appendages of subterranean species may depend on the size of the spaces they inhabit (Pipan and Culver 2017). Predictably, an elongation of appendages is an adaptation to life in relatively large subterranean spaces, typically caves (troglomorphism, Zacharda 1979). In contrast, a diminished body size represents an adaptation to life in relatively small, narrow subterranean spaces, typically of soils (edaphomorphism, Zacharda 1979). Altogether, the species collected during this study cover the continuum from troglomorphic to edaphomorphic species. For example, relatively large, long-legged Porrhomma microps (carapace width 0.84 mm) and P. microcavense (carapace width 0.75 mm) represent typical troglomorphic species. They are characteristic of the interface between soil and sandy marlstone that harbors relatively large interspaces (Fig. 2B). Because of harsh climatic conditions during Quaternary glaciations, Central Europe lacks large troglomorphic spiders, known, for example, from Mediterranean caves (Culver et al. 2006). The most troglomorphic species of this region are still small enough to also occur in soil (Růžička 1999).

The edaphomorphic species collected during this study were tiny Centromerus (C. cf. piccolo and C. cf. minutissimus), Porrhomma cambridgei, Iberina microphthalma, and tiny Erigoninae linyphiids, all of which have carapace widths less than 0.6 mm. They probably live in narrow interspaces in soil or loess; however, they also occur in deep caves (Růžička and Buchar 2008; Růžička 2018, 2022).

Different shallow subterranean habitats presumably provide microhabitats that are preferred by different species. Indeed, some subterranean species collected in the soil-bedrock interface seem to exhibit a preference for specific depths and substrates. For example, Centromerus cf. piccolo is characteristic of loess, and Porrhomma microcavense and Iberina microphthalma are characteristic of sandy marlstone. The species that occur in the surface layer of the soil seem not to express such preferences (for example, Centromerus serratus, C. cavernarum, Mioxena blanda, Palliduphantes spp., Porrhomma microphthalmum, P. microps, Tapinocyboides pygmaeus, Cicurina cicur).

The fauna of shallow subterranean habitats seems to be richer in warm areas than in cold ones. The species Centromerus cf. piccolo, Mioxena blanda, Palliduphantes insignis, Porrhomma cambridgei, and Iberina microphthalma are restricted to the warmest regions of Czechia. Additionally, vegetation can be expected to modify the temperature regime in the soil. A large number of the soil species found during this study live in substrates that occur in partly open xerothermic habitats (for example, Centromerus cf. piccolo, Palliduphantes insignis, Tapinocyboides pygmaeus, and Iberina microphthalma). On the other hand, some species live in habitats that maintain stable humidity and temperature, such as scree forests (e.g., Centromerus cavernarum, Porrhomma campbelli, Theonoe minutissima).

Interestingly, the only artificial habitat that was studied, an almost hundred-year-old gravel heap from a polymetallic ore mine near Kostelní Bříza, harbors relatively rich subterranean fauna. Several species with relatively small eyes, in particular Jacksonella falconeri, Pseudomaro aenigmaticus, Tapinocyboides pygmaeus, Centromerus cavernarum, Centromerus cf. minutissimus, Palliduphantes pallidus, Porrhomma campbelli, Theonoe minutissima, Cicurina cicur, and Mastigusa arietina were found here. Thus, at least these species are able to colonize new isolated sites, probably by ballooning, and they cannot be considered relics like the endemic fauna in limestone caves in the Mediterranean. For Pseudomaro aenigmaticus, this ability was confirmed by the capture of specimens in aeroplankton at 12 m height (Blick and Kreuels 2002).