Temperature in Simarrón II was quite stable and the annual averages obtained were 11.7, 11.7, 11.6, 12.0 and 11.9 °C for the sampled stations A, B, C, E and F, respectively (Fig. 5A). Stations A, B, and C correspond to the chamber Via Salas Blancas, and temperatures were slightly colder (~0.3–0.4 °C) than those measured in the stations E and F of Via Salas Negras. The annual fluctuations were 0.13 and 0.19 °C for stations A and B, 0.43 °C for station C, and 0.21 °C for both stations, E and F. The high fluctuations observed in the station C maybe explained by the presence of small cracks at the top of this chamber that connects to the exterior of the cave, as evidenced by organic matter and plant debris found in this area.

Figure 5. 

Temperature (A) and relative humidity (B) data measured at the stations (A–F) for a full cycle in the cave Simarrón II.

The average annual relative humidity (RH) recorded for stations A, B, C, E and F were 81.3%, 82.9%, 86.6%, 90.5% and 91.2% respectively (Fig. 5B). Overall, Via Salas Blancas showed a slightly lower RH (ave = 83.6%) than Via Salas Negras (ave = 90.8%). The annual fluctuations calculated were 9.8%, 10.6% and 7.8% for the sampling stations (A, B, C) in Via Salas Blancas, and 2.0% for stations (E, F) in Via Salas Negras. RH went down to 64.8% during the summer in Via Salas Blancas (station A), while the lowest RH recorded in Vias Salas Negras was 87.4% (station E). The differences between the two areas of the cave are remarkable, and could potentially determine the arthropod-fauna present in each of them. Consistent with a cave located in the Mediterranean basin, humidity decreased in the months of the summer (July-October, dry season) and increased to peak in the months of February-March when the rain and melted snow arrived. In general, during the fall (beginning of the next rainy season of next cycle), the relative humidity increased again to similar levels recorded in previous winter.

A total of 119 specimens of D. mirabilis were collected in this study (without counting the recaptures): 89 were captured in pitfall traps and 30 were direct captures. Fourteen specimens were death or predated by other specimens inside the traps. From the carabids captured alive, 53 were marked by a unique code, five were released without taking any action, and 25 were brought to the laboratory for further biological studies. These were sent to the “Departamento de Ciencias de la Vida” in the “Universidad de Alcala de Henares”, Madrid (Spain) for feeding, ethological and biological studies. Despite that many of the specimens survived under lab conditions for several months, no mating was observed, and no eggs or larvae could be obtained. Very aggressive behavior was detected among specimens of the same and also opposite sex. Interestingly, D. mirabilis uses their mandibles to transport small stones to build a shelter where they remain most of the time. We never observed D. mirabilis feeding in its native habitat (cave Simarrón II), but it was previously reported that North American ground beetles of the genus Promecognathus prey on polydesmid millipedes (Weary and Will 2020). D. mirabilis is a troglobite inhabiting caves exclusively, and millipedes have not been captured in any of the four caves/locations where D. mirabilis has been reported. The most abundant and possible prey of D. mirabilis in the cave Simarrón-II is the isopod (Trichoniscus sp.). Specimens of D. mirabilis kept in the laboratory were fed successfully dipteran larvae and the aforementioned isopod.

From the 53 carabids that were marked, only four were recaptured in this study.

The number of specimens of D. mirabilis observed at different months and stations sampled in the cave fluctuated widely. The highest number of captures were recorded during the winter months. The number of captures dramatically decreased during the summer months, reaching a minimum in September (only month without captures) (Fig. 6A). The most effective trap was pitfall number 8 (32 captures) and followed by 7 (20 captures), both located in Via Salas Negras. Pitfalls 5 and 9 followed with less than 50% of the captures of pitfall 8 (Fig. 6B). Traps 1 (1 capture) and 2 (no captures) had the least amount of captures, both located in Via Salas Blancas (Fig. 6B).

Figure 6. 

Total number of captures of Dalyat mirabilis per month (A) and per pitfall trap (B) during the period of the study the cave Simarrón II.

We explored the possible relationship of the distance of each pitfall trap location to the entrance of the cave and the number of carabid captures obtained per trap (Fig. 7). A direct and positive correlation was found, the highest number of captures were in pitfalls farther away from the entrance (stations F, E and C) (Fig. 7). Pitfall traps 1, 2 and 3 located in stations A and B were the closest to the entrance of the cave and also with the least amount of beetle captures (Fig. 7).

Figure 7. 

Box-plot representing the abundance and variability of Dalyat mirabilis in function of the distance to the cave entrance along each sampling station.

In general, an estimation of the population size of soil arthropods based on using pit-fall traps alone is difficult because it is affected by several factors, including how motile the species subject of the study is. However, several pit-fall traps placed strategically on a grid could be used to get an approximation to it (Raworth and Choi 2001). Mathematical models could be used to estimate the population size using properly spaced pit-fall traps distributed using a spatial system in crossed axes (Perner and Schueler 2004). The main problem for the design of such a system in a small/medium sized cave is that the possible location of the traps is limited by the development of the cave and the human ability to progress through it. Tercafs and Brouwir (1991) used this method successfully to estimate the population of other cave-dwelling beetles.

Simarrón II has narrow, tortuous and complicated paths, and it is made of large blocks and little or no substrate which makes the placement of traps extremely difficult. Perhaps, this has determined the low number of recaptures in this study, only four specimens in the entire sample period. The low number of beetles recaptured invalidate the application of any formula/models used previously to determine population size, and even if applied, the results would not be relevant using such a low recaptured numbers. It was not optimal that several individuals (14) were found dead in the traps, some live specimens were taken for laboratory experiments (25) and that not all individuals captured alive were marked (5). An observation we made of this species is that it is extremely territorial and aggressive. This may explain the presence of so many dead individuals in the traps (usually body parts); it is a very confined space for several territorial individuals trapped.

There is an important presence of this species in Salas Negras, this is justified by the relatively high number of specimens captured in this part of the cave. However, as mentioned previously, the number of recaptures is low. Other than the explanations given above for this low number, it is also possible that the population size is relatively big, and therefore new “non-marked” specimens are always captured in each sampling trip. This will also imply that the specimens do not “hang around” the area where the trap is located since they are not recaptured, and instead, they keep moving within the interstice of the cave (they are not seeing again). It is an unlikely possibility that marked and released specimens learn to do not fall again in the traps. In any case, future studies will address some problems encountered here and will aim to determine the population size of this enigmatic species.

Several spatially-resolved analyses were performed to understand the distribution of D. mirabilis in the cave Simarrón II. A distribution map integrating relative humidity, temperature, and the number of captures per trap was created (Fig. 8). This map indicate that the most abundant captures happened in areas of the cave with high relative humidity (>82%), and air temperatures between 11.25 °C and 12.25 °C (Fig. 8). This may indicate that D. mirabilis prefers the inner, and most climatically stable areas of the cave, in agreement with the general trend that species richness and diversity of troglobionts are typically greatest deeper inside caves (Kozel et al. 2019).

Figure 8. 

Distribution map for the number of specimens of Dalyat mirabilis depending on the cave air temperature and relative humidity. Estimations of the number of specimens are based on the sampling (captures) and temperature and relative humidity records at each station (labelled as A, B, C, E and F) during an annual cycle, and using the inverse to a distance gridding calculation.

To further understand the influence of each variable, we analyzed independently whether strong fluctuations in the temperature or relative humidity determine the presence/absence or abundance of D. mirabilis. In a first analysis, a strong positive correlation was found for pitfalls with a high number of D. mirabilis captured and environmental conditions with a low coefficient of variation (CV) of the relative humidity at those sampling stations. Those stations with the highest number of captures were in areas of the cave with the lowest fluctuations of relative humidity (Fig. 9). We could not find a clear correlation when a second analysis was carried out using the same captures per station and the CV (%) of the cave air temperature. Stations with the highest and also lowest number of captures had relatively stable temperature; however, station C that has the highest annual variation in the air temperature recorded more specimens than station A and B with lowest number of captures but more stable temperature (Fig. 10).

Figure 9. 

Bubble plot of the presence and abundance of Dalyat mirabilis as a function of the coefficient of variation (CV, %) for the relative humidity. The coefficient of variation represents the percentage of the standard deviation to the mean, and it is a useful statistic for comparing the degree of microclimatic stability (in terms of time-variation of humidity) from one location to another into the cave. The size of the blue circles indicates the relative abundance of Dalyat mirabilis. The dashed linear function (slope: -0.87, y-axis intercept: 10.20 and R² = 0.81) shows a remarkable increment of presence and abundance of Dalyat mirabilis as the CV of relative humidity decreases below 3%.

Figure 10. 

Bubble plot of the presence and abundance of Dalyat mirabilis as a function of the coefficient of variation (CV, %) for the cave air temperature. The coefficient of variation represents the percentage of the standard deviation to the mean, and it is a useful statistic for comparing the degree of microclimatic stability (in terms of time-variation of temperature) from one location to another into the cave. The size of the blue circles indicates the relative abundance of Dalyat mirabilis.

Pitfall traps located in Salas Negras away from the entrance showed more stable environmental conditions: higher and more stable relative humidity compared to Salas Blancas, higher number of captures (% of the total) and a slightly higher and stable temperature (Figs 5, 8, 9 and 10). In fact, this is the only area of the cave that the substrate and walls does not dry out in the summer months; and the environmental conditions (humidity and temperature) favored the presence of D. mirabilis all year round. These results suggest that the presence of D. mirabilis maybe determined mostly by a high and relatively stable relative humidity. This is in line with some early studies on the distribution of cave-inhabiting terrestrial arthropods. It was reported that most abundant populations are usually a short distance inside the dark zone of a cave entrance (Peck et al. 1976). This is where certain microclimatic stability is reached, i.e. contrary to what might be expected, which would be to find a greater species diversity and larger population sizes in the most external parts of the cave with the greatest food availability. A more stable cave environment also favors some species over others; especially species with high specialization for cave life, or edaphic spaces (Tobin et al. 2013; Pellegrini et al. 2016). Other studies suggest that the distribution of several epigeous species (trogloxenes or accidentals) is also linked to cave features (e.g. cave morphology and humidity conditions, among others), and follows patterns that are similar to the ones observed for more usual cave dwellers (Lunghi et al. 2014). Since the temperature was more or less stable across the cave (Figs 2, 5), more studies with a broader range of temperatures are needed to determine the preference of D. mirabilis, and its tolerance to seasonal fluctuations.

The invertebrate fauna that Simarrón II hosts is unique and with a high level of endemism. A total of 31 invertebrate species were identified in this cave and two are in the process of being described as new species (Table 1). The 31 species includes 8 (25.8%) obligate cave dwellers (troglobites), 9 (29.0%) facultative cave dwellers (troglophiles) and 14 (45.2%) are animals found in caves, but also on the surface, where they encounter proper conditions for their life cycles (trogloxenes).

In addition to D. mirabilis, there is another coleopteran of the family Carabidae, Platyderus speleus, captured in Simarrón II. This species was originally described from the touristic Caves of Nerja in south Spain (Malaga). It has also been captured in three other caves in the province of Almería (data not published). This is an emblematic species since it is one of two species present in Spain of the genus Platyderus that shows a high adaptation to live in caves (Zaballos and Jeanne 1994), although it can be found outside caves under big rocks too (Anichtchenko 2009).

Potential preys of these two carabids are 8 species of collembolans captured in Simarrón II (Martínez et al. 2004). They are edaphic and troglophiles and distributed along the cave. Some of the species, Arrhopalites pygmaeus or Gisinurus malatestae are commonly found in the soil in Europe or in Mediterranean countries. Yoshiiphorura bellingeri is distinct from other species known and represents a new genera and species described from specimens captured in three caves located at high altitude in Sierra de Gádor (all of them over 1,600 m.a.s.l.). It is a troglophile species, not associated with guano, and it was captured in pitfall traps during all year; however, the number of specimens captured in summer was considerably lower than in the other months.

Two singular species of pseudoscorpions, Chthonius mayorali and Neobisium piquerae, were captured in Salas Blancas. Ch. mayorali was associated with dryer areas in this big chamber, while N. piquerae occupied soil and substrate that is permanently wet. In the winter, Ch. mayorali was only found in the upper part of the chamber Salas Blancas flowstone, probably displaced by N. piquerae, that is normally spotted and captured in the lower portion of this space. They are both endemic of Sierra de Gádor and C. mayorali has being captured in two other caves, and N. piquerae has been reported only from one more cave (Cueva de la Corraliza) where both species also coexist.

The isopod Trichoniscus sp. is very abundant in the deepest chamber of Salas Negras all year round, especially in a pile of fresh bat guano where traps 9 and 10 are located. Trichoniscus sp. was captured in traps but also by hand capture. Predation on this species by D. mirabilis has been observed in the laboratory. This isopod will be described as a new species.

The only identified spider, Tegenaria aff. herculea shows troglomorphic adaptations, such as eye size reduction, elongation of appendages and depigmentation (Ribera et al. 2000).

There are two other endemic species captured in Simarrón II that are widespread in other caves in Sierra de Gádor. The first one is the cricket Petaloptila barrancoi that has been captured in more than twenty caves in these mountains (Barranco and Amate 2008). It is a troglophile species (Barranco and Molina-Pardo 2021) that does not usually go deep in the cave, and it was only found in the chamber Salas Blancas in Simarrón II. The second species is the dipluran Plusiocampa gadorensis, a troglobite species captured along the entire cave of Simarrón II (Sendra et al. 2020). It has also been captured in 10 more caves in Sierra de Gádor: Cueva de la Mudica, Sima Termal, Cueva and Sima del Llano de la Montés, Cueva de la Corraliza, Cueva Nueva, Sima del Aire, Sima del Puntal, Cueva de las Colmenas-II and the artificial water mine Fuente Vieja (data not published).

The flea L. taschenbergi amitina is a common parasite of species of rodents in the genus Apodemus (Gómez et al. 2013); mice have been observed at the bottom of the entrance shaft during our visits to the cave.