Species conservation profiles of endemic spiders (Araneae) from Madeira and Selvagens archipelagos, Portugal

Abstract Background The North Atlantic archipelagos of Madeira and Selvagens present a unique biological diversity including, presently, 56 endemic spider species. Several recent projects provide valuable information on their distribution across most islands and habitats. To date, the only endemic spider assessed according to the IUCN Red List criteria is Hogna ingens. The objective of this paper is to assess all remaining endemic species and advise on possible future conservation actions critical for the survival of endangered species. New information Seven species were found to have a continuing decline in either range or population size. Their decline can be mostly attributed to habitat destruction or degradation, invasive plant species that reduce quality of habitat, forest fires at high mountain regions and possible competition for resources from invasive congeners. The tetragnathid M. barreti is considered as possibly extinct due to the suspected impact of a competing species. Although most endemic spiders from the Madeira and Selvagens archipelagos have relatively low extinction risk due to the good condition and protection of the laurisilva forests where many live, there are a number of species requiring urgent attention and protection measures. These include all cave and mountain-restricted species as well as those threatened by competing congeners or invasive plants. Extending current protected areas, restoring original habitats of threatened species and the control of invasive taxa should remain a priority for species survival.


Introduction
The archipelagos of Madeira and Selvagens are renown for presenting a unique biological diversity, which is a major contributing factor (together with threat levels) for their inclusion in one of the major biodiversity hotspots worldwide, jointly with two other Macaronesian archipelagos (Azores and the Canaries) (Myers et al. 2000). Both geological and geographical factors were important drivers of species richness as well as endemism in the Madeira and Selvagens archipelagos. The geological age of these islands (5-27 My) and their proximity to the mainland or to paleoislands allowed the colonization and diversification of many taxonomic groups (Fernández-Palacios 2010, Fernández-Palacios et al. 2011. Furthermore, the two archipelagos are composed by several islands and islets that share a number of species but also have a considerable number of exclusive species, single-island endemics . The Madeira archipelago comprises three island groups -the Desertas Islands (Deserta Grande - Fig. 1, Bugio and Ilhéu Chão); Porto Santo (including the main island - Fig. 2 -and seven islets); and Madeira proper (the main island - Fig. 3 -and the surrounding islets). The Selvagens archipelago is composed by one island and two islets -Selvagem Grande (Fig. 4), Selvagem Pequena and Ilhéu de Fora. It is however important to highlight the much larger contribution of Madeira Island to the overall species pool (both in terms of species richness and endemism) when compared with the other islands . Madeira Island is the largest, the highest and presents a more diverse set of terrestrial ecosystems, matching the stage of maximum orographic complexity of the volcanic island cycle (Fernández-Palacios et al. 2011).   The most emblematic ecosystem of the archipelago -the laurisilva forest -is restricted to Madeira Island where it covers nearly 20% of the land surface (Menezes et al. 2005, Boieiro et al. 2015. Madeira laurisilva is the largest area of this relict forest and includes some of the most pristine fragments being classified as a World Heritage Site by UNESCO (IUCN 1999), a priority habitat under the Habitats Directive and also benefiting from regional, national and international legislation (Menezes et al. 2005). Both the laurisilva forest and the summit vegetation harbour a large number of endemic species which made them preferential targets of biodiversity and taxonomic studies since the 19th century (Boieiro et al. 2010). The other islands, besides Madeira, are dominated by coastal scrublands and thermophilous woodland (particularly in Porto Santo) and have been less studied.
Recent efforts were made to update the knowledge on the taxonomy and conservation priorities for the archipelagos of Madeira and Selvagens (Martín et al. 2008, Martín et al. 2010. Over 7500 taxa were reported for these archipelagos, of which nearly 20% were endemics, mostly being terrestrial arthropods ). The spider checklist was based on a thorough analysis of the literature plus personal observations and reported the occurrence of 184 species for these archipelagos, including 58 endemics (Cardoso and Crespo 2008). This publication also pointed to some gaps in the knowledge on species taxonomy and distribution that needed to be addressed in future studies. Several recent papers were important contributions to overcome those gaps by reporting new species to the archipelagos and providing valuable information on endemic species distribution (Crespo et al. 2009a, Crespo et al. 2009b, Crespo et al. 2013, Crespo et al. 2014a, Crespo et al. 2014b). Finally, a number of different projects (see acknowledgements) have made important contributions to overcoming these gaps by reporting extensive spider samples across islands and habitats during the last 10 years. These works led to important taxonomic and distribution data changes and we currently know 56 described endemic spider species.
Despite the known vulnerability of many endemic spider species, only the Desertas wolf spider -Hogna ingens (Blackwall, 1857), Fig. 5 -was assessed for extinction risk according to the Red List criteria of the International Union for the Conservation of Nature . This large spider, restricted to Vale da Castanheira in the northern end of Deserta Grande, is threatened by Phalaris spp. grasses which, with their large roots, prevent the spider to access adequate shelters under stones and in soil crevices. A species conservation plan is now underway and includes regular monitoring of the spider population, chemical treatments to control the invasive species and ex situ conservation with possible future reintroduction of adult specimens.
The objective of this paper is to assess the remaining 55 endemic spider species according to the IUCN criteria and advise on possible future conservation actions critical for the survival of endangered species. In the future we intend to assess a number of species currently being described in order to contribute to the proper protection of this unique fauna.

Methods
Species data were collected from all bibliography on Madeira and Selvagens spiders published until July 2017. These included mainly taxonomic and faunistic works. We also used numerous unpublished data collected within multiple projects (e.g. Boieiro et al. 2013, see also acknowledgements) that used standardized sampling (e.g. Cardoso 2009). This new information on species distribution will be published in several forthcoming papers. Whenever possible, with each species record we also collected additional information, namely habitat type and spatial error of coordinates.
For all analyses we used the R package red -IUCN redlisting tools (Cardoso 2017a, Cardoso 2017b. This package performs a number of spatial analyses based on either observed occurrences or estimated ranges. Functions include calculating Extent of Occurrence (EOO), Area of Occupancy (AOO), mapping species ranges, species distribution modelling using climate and land cover, calculating the Red List Index for groups of species, among others. The package also allows the calculation of confidence Hogna ingens (Blackwall, 1857) adult female (photo by Pedro Cardoso).
limits for all measures, an essential but almost invariably forgotten feature in view of unavoidable uncertainty. It outputs geographical range, elevation and country values, maps in several formats and vectorial data for visualization in Google Earth.
In this work, the EOO and AOO were calculated in one of two ways: -for extremely range restricted species for which we assumed to know the full range, these values were classified as observed, the minimum convex polygon encompassing all observations used to calculate the EOO and the 2x2 km cells known to be occupied used to calculate the AOO. When the EOO was smaller than the AOO, it was made equal as per the IUCN guidelines (IUCN Standards and Petitions Subcommittee 2017).
-for widespread species or those for which we did not have confidence to know the full range, we performed species distribution modeling (SDM). This was done based on two environmental datasets depending on the distribution of each species. For single island endemics of Madeira proper (main island) we used 100x100m resolution data on altimetry, slope, annual precipitation, annual maximum and minimum relative humidity, annual maximum and minimum temperature, land cover and a disturbance index based on the latter (Boieiro et al. 2013. For species living on other islands, given the lack of availability of high-resolution data, we used 30 arc-second (approximately 1x1km) resolution data on 19 bioclimatic variables (Fick and Hijmans 2017) reduced to three after performing a PCA (the three first axes) plus land cover.
For SDMs we used ensemble modeling with the Maxent method (Phillips et al. 2006) implemented in red with associated spatial error (used to randomly place records within such error for each run), 100 runs per species and using only a subset of two explanatory variables for each run. Using subsets of explanatory variables was found to output better predictions than using many variables for rare species, with few occurrence records (Lomba et al. 2010, Breiner et al. 2015. Ensembles were summed using the Area Under the Curve (AUC) values to weight each of the 100 runs as: weigth = max(0, (AUC -0.5)) Isolated patches outise the original distribution polygon were then excluded from maps to avoid overestimation of values. When performing SDMs, confidence limits for EOO and AOO were calculated using the number of models (runs) that predicted presence for each cell applying the percentiles 0.025 and 0.975 for the upper and lower limits respectively. All final maps and values were checked and validated by our own expert opinion. KMLs derived from these maps were also produced using the red package. The current threats are believed to cause a decrease in the species population numbers in unknown rates.
-(c) a decline in area of occupancy, extent of occurrence and/or quality of habitat No Yes Yes Unknown The uncontrolled visits by tourists and locals that think of caves as adventure playground, accumulation of litter and use by domestic animals cause major changes in the cave environment and consequent decrease in quality of habitat for the species. This is believed to be leading to a decrease in population numbers, although no monitoring is being made and the rates are unknown.

Decline (inferred)
The quality of habitat is inferred to be decreasing due to severe changes in the environment of caves.
Major Importance -7.1. Caves and Subterranean Habitats (non-aquatic) -Caves Ecology 3 mm 1 No Ecology and traits are largely unknown as the only references to the species are from a short taxonomic description and a checklist. Yet, congeners are known to build sheet webs and this species might build them on cave walls on the hunt for insects. The species was probably driven away from part of its historical range by touristic activities that include digging of new tunnels, water regime modifications (artificial pools) and artificial lighting. Both current locations are threatened by use of caves by domestic animals, uncontrolled visits and accumulation of litter. The population size of this species is unknown. It is possibly threatened by uncontrolled visits to the caves and accumulation of litter (Reboleira et al. 2011) which decrease the habitat quality and ability of the species to occupy its full historical range. This is a troglobiont species with considerable eye reduction, depigmentation and appendage elongation. Ecology and traits are largely unknown as the only reference to the species is from a short taxonomic description. However, congeners are known to build sheet webs and this species might build them on cave walls on the hunt for insects. This species is possibly threatened by uncontrolled visits to the caves and accumulation of litter (Reboleira et al. 2011) which decrease the habitat quality and ability of the species to occupy its full historical range.

Conservation
Needed -1.1. Land/water protection -Site/area protection -1.2. Land/water protection -Resource & habitat protection -2.3. Land/water management -Habitat & natural process restoration -3.2. Species management -Species recovery -5.1. Law & policy -Legislation -5.4. Law & policy -Compliance and enforcement Furnas do Cavalum are considered scientific patrimony by the "Plano Director Municipal" of Machico. Yet, this cave species would benefit from effective protection with adequate legislation of the lava tube system where it occurs with eventual restoration of natural conditions of the environment and recovery of its population. It should also be formally protected by adequate legislation. The taxonomical status of the species and possible synonymy with C. anoculus should be clarified. Research on population trends and its causes is needed to know the real threat levels and how to minimize them. A species conservation plan and a management plan would improve its survival chances for the future.

Habitat (narrative):
Trend in extent, area or quality?: Habitat importance:

Habitats:
Locations 0 No known threats to the species.

Stable
The preferred habitat of the species, humid laurisilva forest, is not experiencing a decline in area and the invasive species present seem not to affect the spider population.

Yes
No population size estimates exist.

Yes
Humid laurisilva forest on the northern slopes of Madeira Island.

Stable
Major Importance -1.9. Forest -Subtropical/Tropical Moist Montane This species is a sheet-web builder on the tree branches and under stones, feeding mainly on small insects. The species seems closely associated to the laurisilva forest.

Threats
Past -12. Other options -Other threat Unknown threats.

Conservation
In Place -1.1. Land/water protection -Site/area protection Most of the species range lies inside the Madeira Natural Park.

Other
International Very important -3.1. Monitoring -Population trends Monitoring of population trends should be conducted to confirm species status.
The preferred habitat of the species, humid laurisilva forest, is not experiencing any decline in area and the invasive species present should not affect the spider population.

Stable
The preferred habitat of the species, humid laurisilva forest, is not experiencing any decline in area and the invasive species present should not affect the spider population.

0
No known threats to the species.

Stable
The preferred habitat of the species, humid laurisilva forest, is not experiencing any decline in area and the invasive species present should not affect the spider population. No population size estimates exist.

Yes
Humid laurisilva forest on the northern slopes of Madeira Island.

Stable
The preferred habitat of the species, humid laurisilva forest, is not experiencing any decline in area and the invasive species present should not affect the spider population.

Region for assessment:
-Global

Geographic range
Biogeographic realm:

Species Distribution Model
Only four records for this species exist, mostly recent and all in laurisilva forest (Denis 1962, Crespo et al. 2014b. It was possible to perform species distribution modeling to predict its potential range with confidence limits. See methods for details.

1850
Dipoenata longitarsis is known from a few sites in laurisilva forest that occupies about 20% of the island, mainly on its steep and humid northern slopes.

Stable
The preferred habitat of the species, humid laurisilva forest, is not experiencing any decline in area and the invasive species present should not affect the spider population.

Stable
The preferred habitat of the species, humid laurisilva forest, is not experiencing any decline in area and the invasive species present should not affect the spider population. Monitoring of population trends should be conducted to confirm species status. Denis, 1962

Region for assessment:
-Global

Geographic range
Biogeographic realm:

Observed
Largely unknown, as there are only two records for the species (Denis 1962, Wunderlich 1992, both on open mountainous areas.

1800
Drassodes rugichelis is known from two sites (Paúl da Serra and Pico do Cidrão), both on open mountain areas. The true range is however unknown and not possible to model with confidence. The ecology of this species is mostly unknown, but it is probable that they are active nocturnal hunters at ground level.

Threats
Ongoing -7.1. Natural system modifications -Fire & fire suppression The mountain areas of Madeira Island have been affected by recent wildfires, which might have affected this species.

Conservation
In Place -1.1. Land/water protection -Site/area protection The distribution of the species should be researched through extensive collections on the islands' mountainous areas above the tree line. Monitoring of population trends should be conducted in the future and the negative effects of possible threats, such as wildfires, should also be assessed.

Region for assessment:
-Global

Geographic range
Biogeographic realm:

-Palearctic
Countries: The species is restricted to the small Selvagens archipelago that lies between Madeira and the Canary Islands, where it possibly occupies all islets -Selvagem Grande, Selvagem Pequena and Ilhéu de Fora (Macías-Hernández et al. 2010). Habitats:

Stable
No current threats to the species.

Stable
No current threats to the species.

Yes
No population size estimates exist.

Stable
No current threats to the species.

Unknown
The Selvagens archipelago is dominated by barren areas with low herbaceous vegetation and rocky outcrops.

Stable
The natural vegetation of Selvagem Grande has been largely recovered by successful projects coordinated by the Madeira Natural Park devoted to the eradication of invasive species.
Major Importance -4.5. Grassland -Subtropical/Tropical Dry -6. Rocky areas (e.g. inland cliffs, mountain peaks) This species is found across the islands living on barren and rocky areas. The diet of D. aneris is unknown, although most congeners are specialized hunters feeding on woodlice. -8.1. Invasive and other problematic species, genes & diseases -Invasive non-native/ alien species/diseases

Future
Although not currently a threat, the invasive D. crocata has been previously found on Selvagem Grande in the past (Macías-Hernández et al. 2010). It is possible that a future re-introduction of the species might lead to competition for resources with D. aneris with unpredictable consequences, as was already suggested for the extinction of at least one endemic Dysdera in the Azores ).

Conservation
In Place -1.1. Land/water protection -Site/area protection The entire range of the species is within the Selvagens Nature Reserve. Monitoring of population trends should be conducted to confirm this species status. In addition, some information on the species life history, namely feeding regimen, should be collected as it might be restricted to very few prey types with implications for its conservation. Denis, 1962 Species information

Region for assessment:
-Global

Geographic range
Biogeographic realm:

Species Distribution Model
Multiple collection sites are recorded for the species, mostly recent and in laurisilva forest. It was possible to perform species distribution modeling to predict its potential range with confidence limits. See methods for details.

1850
Dysdera coiffaiti is known throughout the laurisilva forest that occupies about 20% of the island, mainly on its steep and humid northern slopes. The preferred habitat of the species, humid laurisilva forest, is not declining in area and the invasive species present should not affect this spider's population.

Stable
The preferred habitat of the species, humid laurisilva forest, is not experiencing any decline in area and the invasive species present should not affect this spider's population.

0
No known threats to the species.

Stable
No known threats to the species

Stable
No known threats to the species.

Yes
Humid laurisilva forest on the northern slopes of Madeira Island.

Stable
The preferred habitat of the species, humid laurisilva forest, is not declining in area and the invasive species present should not affect this spider's population.

Geographic range
Biogeographic realm:

Species Distribution Model
A single record is published on the original description without a precise locality (Blackwall 1862). Two more records are recent and to be confirmed and were used on an attemp to model the species distribution.

1850
Dysdera diversa is known only from high altitude areas of laurisilva forest (above 800m).

13-264-644
Stable Inferred to be stable as the preferred habitat area and quality is stable.

12-264-644
Stable Inferred to be stable as the preferred habitat area and quality is stable.

Stable
Possibly stable as the preferred habitat area and quality is stable.

Yes
No species abundance estimates exist

Yes
High altitude laurisilva forest.

Stable
Preferred habitat area and quality is stable. The species is found on few places of the high-altitude laurisilva forest of Madeira Island, living on the soil. The diet of D. diversa is unknown, although most congeners are specialized hunters feeding on woodlice.

Past
-12. Other options -Other threat Unknown threats.

Conservation
In Place -1.1. Land/water protection -Site/area protection All the species range is inside the Madeira Natural Park.

Region for assessment:
-Global

Geographic range
Biogeographic realm:

Observed
Only two subpopulations of this species are known. These are in close proximity, on the Island of Porto Santo, which has been extensively surveyed, therefore the entire current distribution of this species should be known.

320
The species seems to be restricted to the northeastern part of the island of Porto Santo. It was originally described from Pico Branco (Wunderlich 1995), the area with the most extensive and best preserved native vegetation on the island. A single juvenile was recently collected at the top of the neighbouring Pico do Facho, in a small area with few native plants (e.g. Heberdenia excelsa) within exotic pine and cedar trees. Although the species is not known to ever have occupied areas outside its current range, this is extremely small (EOO is in effect < 1km ) and restricted to two nearby peaks with very few remnant native vegetation, with most areas surrounding them being converted to agricultural fields now abandoned or exotic pine and cedar plantations.

Area of occupancy
The species is not known to ever have occupied areas outside its current range, this being extremely small (AOO is in effect < 1km ) and restricted to two peaks with very few remnant native vegetation. Most areas surrounding them were converted to agricultural fields now abandoned or exotic pine and cedar plantations.

2
The two peaks where the species is found are surrounded by exotic tree plantations with numerous invasive plant species. The species seems to be able to survive mostly among native vegetation but also in few sheltered sites with planted trees. Yet, the spread of invasive plants might jeopardize the subpopulations in these two peaks. Further subpopulations were almost certainly lost in nearby peaks (e.g. Moledo or Pico da Gandaia), but probably before the species description in 1995.

Population
Unknown Decline (inferred) Inferred from possible decline in habitat quality (leading also to possible decline in EOO and AOO) due to the effects of invasive plant species that do not provide adequate shelter for the spider.
-(c) a decline in area of occupancy, extent of occurrence and/or quality of habitat

Yes
No population size estimates exist.

Yes
As only one juvenile individual (identified through clear somatic characters) was found in Pico do Facho despite intensive sampling, it is possible that this subpopulation is endangered due to loss of habitat quality in the near future. That would mean that only a single subpopulation (50%) in Pico Branco would be left.

Yes
The species seems to be restricted to rocky or few native vegetation areas with Erica platycodon or Heberdenia excelsa often within exotic pine and cedar.

Conservation action type:
Possible estimated decline in habitat quality (leading also to possible inferred decline in EOO and AOO) due to the effects of invasive plant species that do not provide adequate shelter for the spider.
Major Importance -3.8. Shrubland -Mediterranean-type Shrubby Vegetation -6. Rocky areas (e.g. inland cliffs, mountain peaks) The species is found living within scarce patches of native vegetation. The diet of D. portisancti is unknown, although most congeners are specialized hunters feeding on woodlice.

Threats
Ongoing -8.1. Invasive and other problematic species, genes & diseases -Invasive non-native/ alien species/diseases As the species seems to be able to survive mostly among native vegetation, probably due to difficulty in finding adequate shelter within other plant species, the spread of invasive plants might jeopardize its survival.

Conservation
In Place -1.1. Land/water protection -Site/area protection Needed Part of the original habitat (Pico Branco) is included in the Natura network, but both localities urgently need to be restored jointly with the neighbouring peaks. The spider would benefit from recovery and re-introduction to these new areas which should have been part of its historical range with possible ex-situ breeding for both re-introduction and raising awareness on its emperiled status. Monitoring of population and habitat trends should be conducted to confirm species status. In addition, some information on life history, namely feeding regimen, should be collected about the species, as it might be restricted to very few prey types with implications for its conservation. The species would benefit from a species conservation plan that would include recovery actions for both the spider and the habitat and a management plan for new protected areas to be created within its historical range.

Geographic range
Biogeographic realm:

-Portugal
Suppl. material 13 Unknown Distribution of the species is unknown as there is only a single record from Caldeirão do Inferno in 1962 (Denis 1962). It has not been found since this despite recent survey efforts. 1500 1500 Distribution of the species is unknown as there is a single record from Caldeirao do Inferno in 1962 (Denis 1962). The diet of D. vandeli is unknown, although most congeners are specialized hunters feeding on woodlice.

Past
-12. Other options -Other threat Unknown threats.

Conservation
In Place -1.1. Land/water protection -Site/area protection The known species range is inside the Madeira Natural Park. This species has not been found since the original description (Denis 1962) and it needs, first of all, taxonomic clarification. If valid, basic information would be needed on its distribution, ecology and possible threats. Not recorded since original description from Madeira with uncertain locality (Kulczyński 1899).

Region for assessment:
-Global

Geographic range
Biogeographic realm: The species EOO and AOO are unknown.
Only mentioned from Madeira Island, with no locality data. The species has not been found since original description in 1899 (Kulczyński 1899) and needs, first of all, taxonomic clarification. If valid, basic information would be needed on its distribution, ecology and possible threats. was possible to perform species distribution modeling to predict its potential range with confidence limits. See methods for details.

1450
Frontinellina dearmata is known throughout the laurisilva forest that occupies about 20% of the island, mainly on its steep and humid northern slopes.

Stable
The preferred habitat of the species, humid laurisilva forest, is not experiencing any decline in area and the invasive species present should not affect the spider population.

Stable
The preferred habitat of the species, humid laurisilva forest, is not experiencing any decline in area and the invasive species present should not affect the spider population.

Geographic range
Biogeographic realm:

Species Distribution Model
Multiple collection sites have been recorded for the species, mostly recent and in laurisilva forest (Schenkel 1938, Denis 1962, Wunderlich 1987, Crespo et al. 2014b. It was possible to perform species distribution modeling to predict its potential range with confidence limits. See methods for details.

1550
Frontiphantes fulgurenotatus is known throughout the laurisilva forest that occupies about 20% of the island, mainly on its steep and humid northern slopes.

Stable
The preferred habitat of the species, humid laurisilva forest, is not experiencing any decline in area and the invasive species present should not affect the spider population. The preferred habitat of the species, humid laurisilva forest, is not experiencing any decline in area and the invasive species present should not affect the spider population.

0
No known threats to the species.

Stable
The preferred habitat of the species, humid laurisilva forest, is not experiencing any decline in area and the invasive species present should not affect the spider population.

Yes
No population size estimates exist. The preferred habitat of the species, humid laurisilva forest, is not experiencing any decline in area and the invasive species present should not affect the spider population.

Region for assessment:
-Global

Geographic range
Biogeographic realm:

Species Distribution Model
Multiple collection sites are recorded for the species, mostly recent and in laurisilva forest (Schenkel 1938, Denis 1962, Crespo et al. 2014b. It was possible to perform species distribution modeling to predict its potential range with confidence limits. See methods for details.

1700
Hahnia insulana is known throughout the laurisilva forest that occupies about 20% of the island, mainly on its steep and humid northern slopes. The preferred habitat of the species, humid laurisilva forest, is not experiencing any decline in area and the invasive species present should not affect the spider populations.

Yes
No population size estimates exist.

Habitat
Terrestrial Yes Humid laurisilva forest on the northern slopes of Madeira Island.

Stable
The preferred habitat of the species, humid laurisilva forest, is not experiencing any decline in area and the invasive species present should not affect the spider populations. Despite intensive searches during the last decade it was not possible to find this species recently described from undetermined locality in the island of Porto Santo (Wunderlich 1992). It probably is a junior synonym of Hogna insularum.

Region for assessment:
-Global

Geographic range
Biogeographic realm:

Unknown
The species EOO and AOO are unknown.
Only recorded from the Island of Porto Santo, from undetermined locality.

Region for assessment:
-Global

Geographic range
Biogeographic realm:

Species Distribution Model
Multiple collection sites are recorded for the species, many of them recent and in a number of different habitats (Thorell 1875, Warburton 1892, Simon 1897, Kulczyński 1899, Schenkel 1938, Denis 1962, Denis 1963, Wunderlich 1992, Crespo et al. 2013. It was possible to perform species distribution modeling to predict its potential range with confidence limits. See methods for details.

1755
Hogna heeri is known from varied and contrasting habitats, from open barren areas in Bugio (Desertas) to laurisilva forest in Madeira Island.

876-1087-1439
Stable The species seems to be able to live on all kinds of habitat, even close to human settlements.

Stable
The species seems to be able to live on all kinds of habitat, even close to human settlements.

Yes
No population size estimates exist. Multiple collection sites have been recorded for the species, mostly recent in all Madeiran islands at low altitude (Kulczyński 1899, Denis 1962, Denis 1963, Schmidt 1990, Wunderlich 1992, Crespo et al. 2013. It was possible to perform species distribution modeling to predict its potential range with confidence limits. See methods for details.  Open grassland or shrubland with frequent rock outcrops, very common in the eastern part of the archipelago at low altitudes.

Stable
The preferred habitat of the species is not experiencing any decline in area and the invasive species present should not affect the spider populations.
Major Importance -3.8. Shrubland -Mediterranean-type Shrubby Vegetation -4.5. Grassland -Subtropical/Tropical Dry -6. Rocky areas (e.g. inland cliffs, mountain peaks) Ecology 11-20 mm 1 No Active ground hunter feeding mainly on small/medium size arthropods. Multiple collection sites are recorded for the species, usually in open areas but at all altitudes, from coastal areas to the plateau of Paúl da Serra (Walckenaer 1837, Johnson 1863, Thorell 1875, Warburton 1892, Simon 1897, Kulczyński 1899, Schenkel 1938, Roewer 1960, Denis 1962, Denis 1963, Wunderlich 1992, Wunderlich 1995. It was possible to perform species distribution modeling to predict its potential range with confidence limits. See methods for details.

304-368-908
Stable The species seems to be able to live on several habitat types, even close to human settlements.

28-368-908
Stable The species seems to be able to live on several habitat types, even close to human settlements. Region for assessment:

Yes
-Global

Geographic range
Biogeographic realm:

Unknown
The species EOO and AOO are unknown.
The true distribution of Hogna nonannulata is unknown (Roewer 1960, Wunderlich 1995, although it might be present exclusively in laurisilva forest. Multiple collection sites are recorded for the species (Blackwall 1857, Johnson 1863, Schmitz 1895, Kulczyński 1899, Cockerell 1924, Denis 1962, Schmidt 1990, Wunderlich 1992. It was possible to perform species distribution modeling to predict its potential range with confidence limits. See methods for details. 0 270 Across the entire island of Porto Santo and its small islets. Habitats include areas close to human settlements but exclude densely forested areas.

32-64-68
Stable The species seems to be able to live on several habitat types, even close to human settlements.

No
The species seems to be able to live on several habitat types, even close to human settlements. It only avoids densely forested areas.

Threats
Past -12. Other options -Other threat Unknown threats.

Conservation
In Place -1.1. Land/water protection -Site/area protection A small part of the species range is inside the Porto Santo Network of Marine Protected Areas (which include islets). Multiple collection sites are recorded for the species, mostly recent and in all habitat types (Blackwall 1862, Kulczyński 1899, Schenkel 1938, Denis 1962, Wunderlich 1992, Crespo et al. 2013, Crespo et al. 2014b. It was possible to perform species distribution modeling to predict its potential range with confidence limits. See methods for details. The species seems to be able to live on several habitat types, even close to human settlements.

Yes
No population size estimates exist.

No
The species seems to be able to live on several habitat types, even close to human settlements.

1850
Lepthyphantes impudicus is known throughout the laurisilva forest that occupies about 20% of the island, mainly on its steep and humid northern slopes.

Stable
The preferred habitat of the species, humid laurisilva forest, is not experiencing a decline in area and the invasive species present should not affect the spider populations. Multiple collection sites are recorded for the species in laurisilva forest (Schenkel 1938, Wunderlich 1987. It was possible to perform species distribution modeling to predict its potential range with confidence limits. See methods for details.

1850
Lepthyphantes lundbladi is predicted to be present throughout the laurisilva forest that occupies about 20% of the island, mainly on its steep and humid northern slopes. The preferred habitat of the species, humid laurisilva forest, is not experiencing any decline in area and the invasive species present should not affect the spider populations.

Yes
No population size estimates exist. Multiple collection sites are recorded for the species, mostly recent and in laurisilva forest (Wunderlich 1992, Crespo et al. 2014b. It was possible to perform species distribution modeling to predict its potential range with confidence limits. See methods for details.

1850
Lepthyphantes mauli is known from different sites in laurisilva forest that occupies about 20% of the island, mainly on its steep and humid northern slopes.

Stable
The preferred habitat of the species, humid laurisilva forest, is not experiencing any decline in area and the invasive species present should not affect the spider populations.

Geographic range
Biogeographic realm:

Species Distribution Model
Multiple collection sites have been recorded for the species in both Porto Santo and Desertas (Wunderlich 1992, Crespo et al. 2013. It was possible to perform species distribution modeling to predict its potential range with confidence limits. See methods for details.

Stable
The species seems to be common in open areas of several islands and islets.

Yes
No population size estimates exist.

Terrestrial No
Macaroeris desertensis is known throughout the islands and islets of Porto Santo and Desertas, mainly on open grassland, scrubland and rocky areas.

Stable
The species seems to be common in open areas of several islands and islets. Multiple collection sites are recorded for the species, mostly recent and in laurisilva forest (Kulczyński 1899, Denis 1962, Crespo et al. 2014b. It was possible to perform species distribution modeling to predict its potential range with confidence limits. See methods for details.

1850
Macarophaeus cultior is known throughout the laurisilva forest that occupies about 20% of the island, mainly on its steep and humid northern slopes.

Ecology and traits (narrative):
The preferred habitat of the species, humid laurisilva forest, is not experiencing any decline in area and the invasive species present should not affect the spider populations.

Yes
No population size estimates exist.

Habitat
Terrestrial Yes Humid laurisilva forest on the northern slopes of Madeira Island.

Stable
The preferred habitat of the species, humid laurisilva forest, is not experiencing any decline in area and the invasive species present should not affect the spider populations. Trend:

Geographic range
Biogeographic realm:

Species Distribution Model
There are only three records for the species, mostly old (Kulczyński 1899) plus a new record from Pico do Cidrão (unpublished). We still performed species distribution modeling to predict its potential range with confidence limits although these should be taken with caution. See methods for details.

650
1850 Known from few sites at high altitude, in laurisilva forest or above treeline.

28-256-720
Decline (inferred) One of the species habitats, humid laurisilva forest, is not experiencing any decline in area and the invasive species present should not affect the spider populations. The other habitat, mountain areas, experienced a severe wildfire in 2010 which might have reduced the species range. One of the species habitats, humid laurisilva forest, is not experiencing any decline in area and the invasive species present should not affect the spider populations. The other habitat, mountain areas, experienced a severe wildfire in 2010 which might have reduced the species population.

No
The only records of the species are from both laurisilva (Queimadas) and the mountain peaks (Pico do Cidrão).

Decline (inferred)
Although the laurisilva areas are extensive and mostly wellpreserved, the mountain areas above the tree-line have suffered recent (2010) extensive wildfires that destroyed much of the native habitat. It is unknown whether these events have affected the species and to what proportion of its range, although this can be inferred.
Major Importance -1.9. Forest -Subtropical/Tropical Moist Montane -6. Rocky areas (e.g. inland cliffs, mountain peaks) Wildfires have negatively impacted the subpopulations above treeline in the past and may do it again in the future.

Conservation
In Place -1.1. Land/water protection -Site/area protection Most of the species range is predicted to be inside the Madeira Natural Park. Given the high uncertainty the species range should be extensively studied. Furthermore, studies on the ecology of the species and how it was affected by wildfire is necessary. Afterwards, monitoring of population and habitat trends should be conducted to confirm species status. Multiple collection sites are recorded for the species mostly in laurisilva forest, although none recently (Warburton 1892, Schmitz 1895, Kulczyński 1899, Bristowe 1925, Bacelar 1937, Schenkel 1938, Denis 1962, Wunderlich 1987, Wunderlich 1992. It was possible to perform species distribution modeling to predict its potential range with confidence limits. See methods for details.

1850
Meta barreti is (or was) known throughout the laurisilva forest that occupies about 20% of the island, mainly on its steep and humid northern slopes.

0-352-899
Decline (inferred) The preferred habitat of the species, humid laurisilva forest, is not experiencing any decline in area. It should be noted, however, that the species has not been recorded for at least two decades despite extensive sampling. At the same time, the endemic congener Meta stridulans is now commonly seen after description in 1987, about the same time as M. barreti was last recorded. It is possible there was replacement of one species by the other, both single island endemics, although this is for now only suspected. The preferred habitat of the species, humid laurisilva forest, is not experiencing any decline in area. It should be noted, however, that the species has not been recorded for at least two decades despite extensive sampling. At the same time, the endemic congener M. stridulans is now commonly seen after description in 1987, about the same time as M. barreti was last recorded. It is possible there was replacement of one species by the other, both single island endemics, although this is for now only suspected.

0-1
The species is not recorded for at least two decades despite extensive sampling. At the same time, the endemic congener M. stridulans is now very common in the same habitat and region. If this latter species replaced M. barreti it was a fast single event, although this can only be suspected.

Decline (inferred)
The preferred habitat of the species, humid laurisilva forest, is not experiencing any decline in area. It should be noted, however, that the species is not recorded for at least two decades despite extensive sampling. At the same time, the endemic congener M. stridulans is now commonly seen after description in 1987, about the same time as M. barreti was last recorded. It is possible there was replacement of one species by the other, both single island endemics, although this, for now, is only suspected.

Decline (inferred)
The preferred habitat of the species, humid laurisilva forest, is not experiencing any decline in area. It should be noted however that the species is not recorded for at least two decades despite extensive sampling. At the same time, the endemic congener M. stridulans is now commonly seen after description in 1987, about the same time as M. barreti was last recorded. It is possible there was replacement of one species by the other, both single island endemics, although this, for now, is only suspected.

Yes
Humid laurisilva forest on the northern slopes of Madeira Island.

Decline (inferred)
The preferred habitat of the species, humid laurisilva forest, is not experiencing any decline in area. It should be noted, however, that the species is not recorded for at least two decades despite extensive sampling. At the same time, the endemic congener M. stridulans is now commonly seen after description in 1987, about the same time as M. barreti was last recorded. It is possible there was replacement of one This species is an orb-web builder, possibly on the tree branches, feeding mainly on small insects. The congener M. stridulans is much larger (up to 11 mm), possibly giving it a competitive advantage.

Ongoing -8.2. Invasive and other problematic species, genes & diseases -Problematic native species/diseases
The endemic congener M. stridulans is now commonly seen after description in 1987, about the same time as M. barreti was last recorded. It is possible there was replacement of one species by the other, both single island endemics, although this is for now only suspected.

Conservation
In Place

Basis of EOO and AOO:
All the species range is inside the Madeira Natural Park. If its apparent disappearance from the native range is confirmed some measures targeting species recovery should be implemented. The true distribution of the species and possible confusion with Meta stridulans should be clarified. Also research on the possible temporal replacement between these two species is needed. Multiple collection sites are recorded for the species, mostly recent and in laurisilva forest (Wunderlich 1987, Crespo et al. 2014b. It was possible to perform species distribution modeling to predict its potential range with confidence limits. See methods for details.

1400
Meta stridulans is known throughout the laurisilva forest that occupies about 20% of the island, mainly on its steep and humid northern slopes.

Stable
The preferred habitat of the species, humid laurisilva forest, is not experiencing any decline in area and the invasive species present should not affect the spider populations. No known threats to the species.

Stable
The preferred habitat of the species, humid laurisilva forest, is not experiencing any decline in area and the invasive species present should not affect the spider populations.

Yes
No population size estimates exist.

Habitat
Terrestrial Yes Humid laurisilva forest on the northern slopes of Madeira Island.

Region for assessment:
-Global

Geographic range
Biogeographic realm:

Unknown
The species EOO and AOO are unknown.
Misumena nigromaculata is known only from Funchal, south coast of Madeira Island, captured in October 1940 (Denis 1963) and was recently found with doubts in identification at Dunas da Piedade, Ponta de São Lourenco (unpublished). The only confirmed locality and location is now mostly residential area in the outskirts of Funchal.

. Conservation Planning -Area-based Management Plan
The true distribution of the species should be assessed along with possible threats along its range. If endangered, a species conservation plan encompassing recovery actions and area management should be needed to ensure the species survival. The species EOO and AOO are unknown.

Taxonomy
Oecobius minor is known only from the mountainous area south of Faial on the northern coast of Madeira Island, captured in undefined date (Wunderlich 1992). The only other record is from the same island in undefined locality (Kulczyński 1909). Unknown, but congeners build small flat webs over rocks or trunks where they hunt for small insects.

Region for assessment:
-Global

Geographic range
Biogeographic realm:

Observed
The restricted distribution of the species allows to know its EOO and AOO with reasonable confidence. 0 160 The species is probably restricted to the Selvagem Grande Island between Madeira and the Canary Islands, where it was found in an unnamed erosion coastal cave in 1958 (Denis 1963, Rambla 1978, Wunderlich 1992. Two caves are now the entire known range of the species. The cosmopolitan congener O. navus is now commonly seen, even in caves, after being detected on the island for the first time before 1978. It is possible there is a gradual replacement of one species by the other, although this is for now only suspected.

1
If the suspicion that the introduction of O. navus is affecting O. selvagensis is confirmed, there is a single location derived from the spread of the invasive species in a single event.

Stable
Probably stable for the last 10 years.

Decline (inferred)
The cosmopolitan congener O. navus is now commonly seen, even in caves, after being detected on the island for the first time before 1978. It is possible there is a gradual replacement of one species by the other, although this is for now only suspected. It is probable that this is a species that lives underground due to the two known localities being inside of caves and it being largely depigmented. Congeners build small flat webs where they hunt for small insects.

Threats
Ongoing -8.1. Invasive and other problematic species, genes & diseases -Invasive non-native/ alien species/diseases The cosmopolitan O. navus may be outcompeting the species in part of its range.

Conservation
In Place -1.1. Land/water protection -Site/area protection Needed -2.2. Land/water management -Invasive/problematic species control -3.2. Species management -Species recovery -3.3. Species management -Species re-introduction -3.4. Species management -Ex-situ conservation The island of Selvagem Grande is part of the Selvagens Nature Reserve. If competition is confirmed, the invasive O. navus should be controlled. As this task is probably impossible, ex-situ conservation with eventual reintroduction and recovery might be the only feasible measure to prevent further reduction of O. selvagensis. The species is not found since original description in 1899 (Kulczyński 1899) and needs, first of all, taxonomic clarification to confirm synonymy with P. nemoralioides. If valid, basic information would be needed on its distribution, ecology and possible threats.

Region for assessment:
-Global

Geographic range
Biogeographic realm:

Species Distribution Model
Multiple collection sites are recorded for the species, mostly recent and in laurisilva forest (Kulczyński 1905, Schenkel 1938, Denis 1962, Crespo et al. 2014b. It was possible to perform species distribution modeling to predict its potential range with confidence limits. See methods for details.

1750
Philodromus insulanus is known throughout the laurisilva forest that occupies about 20% of the island, mainly on its steep and humid northern slopes.

Stable
The preferred habitat of the species, humid laurisilva forest, is not experiencing any decline in area and the invasive species present should not affect the spider populations. The preferred habitat of the species, humid laurisilva forest, is not experiencing any decline in area and the invasive species present should not affect the spider populations.

Yes
No population size estimates exist.

Yes
Humid laurisilva forest on the northern slopes of Madeira Island.

Stable
The preferred habitat of the species, humid laurisilva forest, is not experiencing any decline in area and the invasive species present should not affect the spider populations.

No
The species is an active hunter at low to high vegetation feeding mainly on small size arthropods.

Threats
Past -12. Other options -Other threat Unknown threats.

Conservation
In Place -1.1. Land/water protection -Site/area protection Most of the species range is inside the Madeira Natural Park.

Region for assessment:
-Global

Geographic range
Biogeographic realm:

Unknown
The species EOO and AOO are unknown.

1860
Only known from Pico Ruivo, the highest mountain in Madeira Island with 1861 m altitude (Denis 1962). A single female was captured in April 1957.

Unknown
The single site where the species was sampled is in high mountain above tree-line.

mm 1 No
Nothing is known about the species but congeners are active hunters mainly feeding on small arthropods.

Threats
Ongoing -7.1. Natural system modifications -Fire & fire suppression If this species is restricted to the high peaks of Madeira recent wildfires may have affected its population. Given the lack of information this is purely speculative.

Conservation
In Place -1.1. Land/water protection -Site/area protection If only living in the mountain peaks of Madeira, this species habitat is protected by the Madeira Natural Park. Individuals of this species are not found since the original description (Denis 1962) and it needs, first of all, taxonomic clarification. If valid, basic information would be needed on its distribution, ecology and possible threats.

Number of locations:
Trend:

Number of individuals:
Trend: Causes ceased?:

450
Pholcus dentatus is known from only two sites separated by 25 kms in the northern coast of Madeira Island between 300 and 450 meters elevation, Fonte da Pedra and Ribeira da Janela (Wunderlich 1995). The species ecology is unknown but almost certainly a tangle-web builder feeding on small arthropods. Multiple collection sites are recorded for the species, mostly recent and in laurisilva forest (Wunderlich, 1987, unpublished). It was possible to perform species distribution modeling to predict its potential range with confidence limits. See methods for details.

1850
Pholcus madeirensis was first described from the southern slopes of Madeira, including near Funchal, in undescribed habitat (Wunderlich 1987). More recently it was found in several laurisilva forest sites of the northern slopes.

Stable
The preferred habitat of the species, humid laurisilva forest, is not experiencing any decline in area and the invasive species present should not affect the spider populations.

Stable
The preferred habitat of the species, humid laurisilva forest, is not experiencing any decline in area and the invasive species present should not affect the spider populations.

Stable
The preferred habitat of the species, humid laurisilva forest, is not experiencing any decline in area and the invasive species present should not affect the spider populations.

Yes
No population size estimates exist.

No
Mainly humid laurisilva forest on the northern slopes of Madeira Island, but also known from the southern slopes in undescribed habitat.

Stable
The preferred habitat of the species, humid laurisilva forest, is not experiencing any decline in area and the invasive species present should not affect the spider populations. The ecology of this species is unknown but almost certainly a tangle-web builder feeding on small arthropods.

Threats
Past -12. Other options -Other threat Unknown threats.

Conservation
In Place -1.1. Land/water protection -Site/area protection Part of its range is inside the Madeira Natural Park. Monitoring of population trends should be conducted to confirm species status. As the preferred habitat outside laurisilva forest is unknown, research on distribution and preferred habitats should be a priority. The ecology of the species is unknown but almost certainly a tangle-web builder feeding on small arthropods.

Region for assessment:
-Global

Geographic range
Biogeographic realm:

-Portugal
Suppl. material 42 Observed Four collection sites are recorded for the species, most recently in laurisilva forest close to the northern coast of Madeira (Wunderlich 1987, Wunderlich 1995. It was possible to perform species distribution modeling to predict its potential range with confidence limits. See methods for details.
Pholcus parvus is known from a few sites in laurisilva forest or nearby, always in valleys close to the northern coast of Madeira Island.

Stable
The preferred habitat of the species, humid laurisilva forest, is not experiencing any decline in area and the invasive species present should not affect the spider populations.

Stable
The preferred habitat of the species, humid laurisilva forest, is not experiencing any decline in area and the invasive species present should not affect the spider populations. The preferred habitat of the species, humid laurisilva forest, is not experiencing any decline in area and the invasive species present should not affect the spider populations.

Yes
No population size estimates exist.

Yes
Humid laurisilva forest on the northern valleys of Madeira Island.

Stable
The preferred habitat of the species, humid laurisilva forest, is not experiencing any decline in area and the invasive species present should not affect the spider populations.

No
The ecology of the species is unknown but almost certainly a tangle-web builder feeding on small arthropods.  (Wunderlich 1995). It was possible to perform species distribution modeling to predict its potential range with confidence limits. See methods for details. The ecology of this species is unknown but almost certainly a tangle-web builder feeding on small arthropods.

Region for assessment:
-Global

Species Distribution Model
Only three collection sites have ever been recorded for this species, mostly from laurisilva forest (Blackwall 1862, Schenkel 1938, Wunderlich 1995. It was possible to perform species distribution modeling to predict its potential range with confidence limits, although it must be carefully considered due to the low number of samples. See methods for details.

1850
Prinerigone pigra seems to be restricted to high-altitude forest (above 600 m) in Madeira Island.

52-456-87
Unknown There are no recent collections of the species. It does not seem to be common given the scarce number of records. Impossible to infer on the current trend.

Terrestrial
Unknown It seems to be mostly found in high-altitude laurisilva forest but possibly also in other habitats above 600m.

Unknown
There are no recent collections of the species. It does not seem to be common given the scarce number of records. Therefore, it is impossible to infer the current trend. Multiple collection sites are recorded for the species, mostly recent and in laurisilva forest (Schenkel 1938, Denis 1962, Wunderlich 1987, Wunderlich 1992, Crespo et al. 2014b. It was possible to perform species distribution modeling to predict its potential range with confidence limits. See methods for details.

1700
Rugathodes madeirensis is one of the most common species in Madeira Island, known throughout the laurisilva forest and also other habitats including pine plantations and close to urban areas. As the species is thought to be restricted to the three small islands/islets of Selvagens, the EOO and AOO can be calculated with some confidence. 0 160 Restricted to Selvagem Grande, Selvagem Pequena and Ilhéu de Fora in the archipelago of Selvagens.

Stable
The EOO seems to be stable with no signs of loss or known threats.

Stable
The AOO seems to be stable with no signs of loss or known threats. This species is only known from Selvagem Grande (Denis 1963, Wunderlich 1992, Crespo et al. 2009b, being possible to calculate the EOO and AOO with reasonable confidence. The EOO seems to be stable with no signs of loss or known threats.

Stable
The AOO seems to be stable with no signs of loss or known threats.

0
No known threats to the species.

Stable
The population size seems to be stable with no signs of loss or known threats. Possible junior synonym of Steatoda paykulliana (Walckenaer, 1806), a widespread palearctic species (Denis 1962). S. distincta probably is a synonym of a widespread species (Denis 1962) and needs, first of all, taxonomic clarification. If valid, basic information would be needed on its distribution, ecology and possible threats.

Region for assessment:
-Global

Geographic range
Biogeographic realm:

Species Distribution Model
Multiple collection sites are recorded for the species, mostly recent and in laurisilva forest (Schenkel 1938, Denis 1962, Wunderlich 1987. It was possible to perform species distribution modeling to predict its potential range with confidence limits. See methods for details.

1750
Tenuiphantes tenebricoloides is one of the most common species in Madeira Island, known throughout the laurisilva forest and also other habitats including planted/cultivated forest and close to urban areas.

181-488-700
Stable Population Information (Narrative): Although most common in laurisilva forest, the species seems to be able to live on several habitat types, even close to human settlements.

160-488-692
Stable Although most common in laurisilva forest, the species seems to be able to live on several habitat types, even close to human settlements.

0
No known threats to the species.

Stable
Although most common in laurisilva forest, the species seems to be able to live on several habitat types, even close to human settlements.

Yes
No population size estimates exist. Threat type:

No
Most common in humid laurissilva forest but recorded from several habitat types.

Stable
Although most common in laurisilva forest, the species seems to be able to live on several habitat types such as cultivated forest, even close to human settlements.
Major Importance Multiple collection sites are recorded for the species, mostly recent and in laurisilva forest (Wunderlich 1987, Wunderlich 1992, Crespo et al. 2014b. It was possible to perform species distribution modeling to predict its potential range with confidence limits. See methods for details.
Trogloneta madeirensis is one of the most common species in Madeira Island, known throughout the laurisilva forest and also other habitats including caves and close to urban areas.

202-389-764
Stable Although most common in laurisilva forest, the species seems to be able to live on several habitat types, even close to human settlements.

172-368-764
Stable Although most common in laurisilva forest, the species seems to be able to live on several habitat types, even close to human settlements. This tiny spider has been found living among the rocks and trees, shrubs and herbs and also in moss on tree trunks (Wunderlich 1987). It possibly builds three-dimensional webs (Hajer 2000).

Threats
Past -12. Other options -Other threat Unknown threats.

Conservation
In Place -1.1. Land/water protection -Site/area protection Most of the species range is inside the Madeira Natural Park. Multiple collection sites are recorded for the species, mostly recent and in laurisilva forest (Schenkel 1938, Denis 1962, Wunderlich 1987, Crespo et al. 2014b. It was possible to perform species distribution modeling to predict its potential range with confidence limits. See methods for details.

1700
Turinyphia maderiana is known throughout the laurisilva forest that occupies about 20% of the island, mainly on its steep and humid northern slopes.  (Crespo et al. 2013, Crespo et al. 2014b. It was possible to perform species distribution modeling to predict its potential range with confidence limits. See methods for details.

1750
Living on relatively high-altitude open areas across Madeira (Paúl da Serra and the region between the highest mountain peaks), Deserta Grande (south plateau) and Bugio (south and north plateaus).

Region for assessment:
-Global

Geographic range
Biogeographic realm:

-Portugal
Suppl. material 53 Observed Species possibly restricted to the islands of Deserta Grande and Bugio (Wunderlich 1987, Wunderlich 1992, Crespo et al. 2013. The EOO can be calculated with reasonable confidence. 0 400 This species is possibly restricted to the islands of Deserta Grande and Bugio where it is know from three sites but probably extends across the entire islands given the adequate microhabitat (steep slopes). Although the invasive X. nubilus seems to have occupied all the flat areas of Desertas, the steep slopes around the islands may constitute refuge to this single island endemic.

Yes
Locations 1 A single event, the introduction of the invasive species X. nubilus, first detected in 2011, seems to have originated the extirpation of X. grohi from most of the islands. This invasive process may continue in the future to the current refuge of the endemic species, the steep coastal slopes. The introduction of the invasive species X. nubilus, first detected in 2011, seems to have originated the extirpation of X. grohi from most of the islands. This invasive process may continue in the future to the current refuge of the endemic species, the steep coastal slopes.

Justification for conservation actions:
Use type: Ecosystem service type: Research needed:

Justification for research needed:
All of the species range is inside the Desertas Nature Reserve. The invasive X. nubilus should be erradicated from the island. As this task is probably impossible, ex-situ conservation with eventual re-introduction and recovery might be the only feasible measure to prevent the species extinction. The current distribution of the species and possible threats from the invasive congener should be thoroughly studied. X. grohi should be the target of a species conservation plan with consequent area management actions. Monitoring of population trends should be conducted to confirm the species status. (Wunderlich, 1992)  This species is only known from Fajã da Nogueira at the northeastern region of Madeira Island in laurisilva forest (Fig. 3). A single female was recorded in November 1980. As this is an unusual season for most species it is possible the species was missed in recent collections due to its phenology. As all congeners, certainly an ambush hunter feeding on small insects.

Threats
Past -12. Other options -Other threat Unknown threats.

Conservation
In Place -1.1. Land/water protection -Site/area protection The known species range is inside the Madeira Natural Park.  Multiple collection sites are recorded for the species, mostly recent and in open grassland or shrubland (Wunderlich 1987, Wunderlich 1992, Crespo et al. 2009a, Crespo et al. 2009b, Crespo et al. 2013. It was possible to perform species distribution modeling to predict its potential range with confidence limits. See methods for details. 0 160 Known from all islands and few islets of Madeira and Selvagens archipelagos. Always at low altitude in open habitats such as grassland or shrubland.

10682-10682-13343
Stable The species seems to be able to live on several habitat types, even close to human settlements.

32-188-432
Stable The species seems to be able to live on several habitat types, even close to human settlements. No known threats to the species.

Stable
The species seems to be able to live on several habitat types, even close to human settlements.

Yes
No population size estimates exist.

No
Open grassland and shrubland, often in disturbed areas, close to or even inside houses.

Stable
Major Importance

Discussion
Out of 56 species evaluated (Hogna ingens included), there is no reliable information on range and trends for 16 (29%). Among the 40 with reliable information, 29 are widespread, with an estimated EOO > 200 km and AOO > 60 km . Most of these are restricted to the 2 2 laurisilva forest that occupies 20% of the area of Madeira Island and which is well preserved and protected for the most part.
Seven species show a continuing decline in either range or population size. Their decline can be attributed to habitat destruction or degradation (Centromerus anoculus and C. sexoculatus, both exclusive to degraded caves), plant invasive species that reduce the habitat quality (Dysdera portisancti and H. ingens), wildfires at high mountain regions (Mesiotelus maderianus) and possible competition for resources from congeners ( Meta barreti and Xysticus grohi).
The tetragnathid M. barreti is considered as possibly extinct due to the suspected impact of a competing species. Other than habitat destruction, competing species have been found to be the major menace for threatened spiders in Macaronesia .
Competing species are almost always invasive taxa, but can also be native taxa that have spread above natural levels and to new areas due to imbalances in the communities. Such is the possible case of Meta stridulans, only recently described, yet now widespread throughout the native forest. We can also hypothesize that this species is in fact an invasive yet to be found on its original region, although this can only be speculated for now.
A few species are missing critical information to be able to assess their status, namely a couple that are only known from the high peaks of Madeira Island (Drassodes rugichelis and Philodromus simillimus). While all other habitats have been subject to multiple recent projects using standardized intensive sampling, the mountain regions have been sporadically sampled, rarely in recent times. This habitat, known to host multiple endemic species of other taxa such as snails and beetles, was subjected to catastrophic wildfires in 2010 that affected many threatened endemics, and should therefore be the focus for future work with spiders. Furthermore, it will also be important to assess the vulnerability of these mountain habitats and their associated spider fauna to climate change effects.
The effects of climate change, one of the prevailing threats accross the world affecting numerous species, was never studied for Madeira and Selvagens spiders. This is a known threat to Macaronesian bryophytes (Patiño et al. 2016) and Azorean spiders (Ferreira et al. 2016), and might affect many of the taxa assessed here. Yet, no good, high resolution future climatic data exists for Madeira and Selvagens, and a study into this has for now been postponed.
Among all the conservation measures suggested, the restoration of original habitat areas and control of invasive species are often a priority to guarantee the survival of threatened species. Many actions have been undertaken by the local authorities (Instituto das Florestas e da Conservação da Natureza) in Madeira and Selvagens archipelagos to control and eradicate invasive plants and mammals jointly with habitat restoration programs. However, it is important not only to ensure the continuity of those actions, but also to monitor their effectiveness by assessing population changes in selected groups like spiders as they already proved to be efficient and effective bioindicators in many terrestrial ecosystems (Bonte et al. 2002, Scott et al. 2006, including oceanic islands . For a few species (Dysdera portisancti, Meta barreti, Misumena nigromaculata, Oecobius selvagensis and Xysticus grohi), ex-situ conservation, recovery and possible re-introduction could be a last resort or insurance against extinction, if this has not occurred yet.
The network of protected areas in Madeira and Selvagens archipelagos is extensive and covers most of the areas known to harbour higher values of species richness and endemism (Boieiro et al. 2015). A clear gap is however evident in Porto Santo (main island) where some important areas for nature conservation are in urgent need to benefit from legal protection. The selection of new areas for nature conservation in Madeira archipelago must take in consideration the spider fauna, but also other invertebrate groups, since they accomodate the largest fractions of the biological diversity, endemicity and vulnerability to extinction.
In conclusion, although most endemic spiders from Madeira and Selvagens archipelagos are in a favourable situation due to the good condition and protection of the laurisilva forests where many live, there are a number of species requiring urgent attention and protection measures. These include the few cave and mountain-restricted species or threatened by competing congeners or invasive plants. Extending current protected areas, restoring original habitats of threatened species and control invasive taxa are still a priority.  (Wunderlich, 1992) Authors: Cardoso, P. Data type: Distribution Filename: Xysticus grohi (Wunderlich, 1992).kml -Download file (11.25 kb) Suppl. material 54: Distribution of Xysticus madeirensis (Wunderlich, 1992) Authors: Cardoso, P. Data type: Distribution Filename: Xysticus madeirensis (Wunderlich, 1992).kml -Download file (4.13 kb)