Terminology relating to adult morphology largely follows Hita Garcia and Fischer (2014) as used by Mbanyana et al. (2018) with some differences as discussed below; descriptions of surface sculpture follow Harris (1979). All measurements (in millimetres) and indices are presented as minimum and maximum, with arithmetic means in parentheses. Abbreviations of the measurements taken and the indices based on them are as follows:

Measurements

EL Eye length: maximum diameter of the compound eye measured in oblique lateral view;

HL Head length: length of head measured in full-face view, from the midpoint of the clypeal margin to the midpoint of the occipital margin; where either of these margins is concave the measurement is taken from the midpoint of a line joining the anterior-most portions of the clypeus or the posterior-most portions of the occipital margin so that the impressions on these margins do not reduce HL;

HL2 Head length 2: maximum distance from the midpoint of the anterior clypeal margin to the midpoint of the posterior margin of head, measured in full-face view; impressions on the anterior clypeal margin and the posterior head margin reduce HL2;

HW Head width: width of the head directly behind the eyes measured in full-face view;

PH Pronotal height: maximum height of the pronotum measured in lateral view;

PPH Postpetiole height: maximum height of the postpetiole measured in lateral view from the highest (median) point of the node to the ventral outline; the measuring line is placed at an orthogonal angle to the ventral outline of the node;

PPL Postpetiole length: maximum length of the postpetiole measured in lateral view;

PPW Postpetiole width: maximum width of the postpetiole measured in dorsal view;

PSL Propodeal spine length: in dorsofrontal view the tip of the measured spine, its base, and the centre of the propodeal concavity between the spines must all be in focus; using a dual-axis micrometer the spine length is measured from the tip of the spine to a virtual point at its base where the spine axis meets orthogonally with a line leading to the median point of the concavity;

PTH Petiolar node height: maximum height of the petiolar node measured in lateral view from the highest (median) point of the node to the ventral outline;

PTL Petiolar node length: maximum length of the dorsal face of the petiolar node from the anterodorsal to the posterodorsal angle, measured in lateral view;

PTW Petiolar node width: maximum width of the dorsal face of the petiolar node measured in dorsal view;

PW Pronotal width: maximum width of the pronotum measured in dorsal view;

SL Scape length: maximum scape length excluding basal condyle and neck;

WL Weber’s length: diagonal length of mesosoma in profile, from the junction of the pronotum and the cervical shield, to the posterior basal angle of the metapleuron.

Indices

CI Cephalic index: HW / HL × 100;

CI2 Cephalic index 2: HW / HL2 × 100;

DMI Dorsal mesosoma index: PW / WL × 100;

DPeI Dorsal petiole index: PTW / PTL × 100;

DPpI Dorsal postpetiole index: PPW / PPL × 100;

LMI Lateral mesosoma index: PH / WL × 100;

LPeI Lateral petiole index: PTL / PTH × 100;

LPpI Lateral postpetiole index: PPL / PPH × 100;

OI Ocular index: EL / HW × 100;

PeNI Petiolar node index: PTW / PW × 100;

PPI Postpetiole index: PPW / PTW × 100;

PpNI Postpetiolar node index: PPW / PW × 100;

PSLI Propodeal spine index: PSL / HL × 100;

PSLI2 Propodeal spine index: PSL / HL2 × 100;

SI Scape index: SL / HW × 100.

Note that 1) HL2 and the CI2 and PSLI2 indices derived from this are included for equivalence with HL, CI and PSLI as reported in Mbanyana et al. (2018), who in following Hita Garcia and Fischer (2014) used for HL the measurement here defined as HL2, 2) petiole node length and postpetiole length are more accurately measured in profile view than in dorsal view as described in Hita Garcia and Fischer (2014), and 3) PeNI and PpNI values as defined here and in Hita Garcia and Fischer (2014) were reported in Mbanyana et al. (2018) but the definitions were omitted. Measurements and indices of all specimens are available in Suppl. material. 1: File S1: Measurements_Tetramorium_nama.xls.

Abbreviations of depositories

AFRC AfriBugs Collection, Pretoria, South Africa

BMNH The Natural History Museum, London, UK

CASC California Academy of Sciences Collection, San Francisco, USA

SAMC Iziko South African Museum Collection, Cape Town, South Africa

The thoroughly illustrated key provided in Mbanyana et al. (2018) contains 18 couplets to distinguish between 19 species; the couplets below are intended to be inserted at the position of couplet 8 in the existing key, with the additional couplet designated as 8a, to allow the remainder of the key to continue unaltered from couplet 9.

Specimens representing three other species within the Tetramorium solidum-group were also found during the 2019 survey of the Oena Diamond Mine; these included three specimens of T. grandinode Santschi, 1913, 113 of T. pogonion Bolton, 1980 and 390 of T. rufescens Stitz, 1923. Tetramorium grandinode and T. rufescens are fairly widespread and are known respectively from at least 22 and 38 other sites in Namibia and South Africa (Bolton 1980; Mbanyana et al. 2018; AntWeb 2020), but the T. pogonion specimens represent only the second record of this species and the first record in South Africa. Tetramorium nama sp. nov. and T. grandinode were rare in the ODM project area and found in only one and two of the 200 pitfall trap samples respectively, while T. pogonion and T. rufescens were relatively common and were found in 22 and 40 samples respectively. The habitat in which the T. nama sp. nov. and T. grandinode specimens were found was open riverine fringe vegetation dominated by two indigenous tree species, Euclea pseudebenus E. Mey. ex A. DC (Black Ebony) and Tamarix usneoides E. Mey. ex Bunge (Wild Tamarisk), although a number of other tree species were present in smaller numbers. Substantial invasion by Prosopis glandulosa Torr. (Honey Mesquite) has occurred in the area and is the subject of an intensive eradication campaign by the Department of Environmental Affairs’ Working for Water (WfW) programme; many cut and poisoned stumps, as well as piles of both dry and freshly cut growth were observed within the survey area (see Figure 3).

Tetramorium nama sp. nov. was found only within the riverine fringe vegetation during the September 2019 ODM survey, but the single record provides insufficient information to indicate whether or not the species might inhabit other areas within the RNP. Two other seed-harvesting species, T. pogonion and T. rufescens, each occurred in five of the six habitat types sampled, together inhabiting all habitat types and suggesting the potential suitability of these habitats for T. nama sp. nov. as well. The absence of any live grasses and apparent dormancy of all non-tree species in Figure 3 highlights the importance of seed storage by T. solidum-group species in ensuring their survival through drought periods. Further, more intensive, surveys would be required to detect habitat preferences of T. nama sp. nov.

The distribution of T. nama sp. nov. beyond the type locality also cannot be determined at this stage, but this species was absent from all of the institutional collections mentioned in Mbanyana et al. (2018) and was not found during any of the targeted surveys that these authors carried out in the Northern Cape and Namibia for T. solidum-group species. No specimens of this species were found during a survey of 48 sites in the Central Namib Desert in 2010 (circa 500 km north of the type locality), nor in several intensive surveys on and around Gamsberg (circa 200 km SE of the type locality) in the Bushmanland Inselberg Region of the Northern Cape from 2009–2017 (unpublished data, AFRC). This suggests that T. nama sp. nov. has a limited distribution, or is very rare within its range, or possibly both. It is however fairly certain that the species will occur in southern Namibia, even if only in habitat similar to where it was found, on the opposite banks of the Orange River.

The future of the Richtersveld National Park

The RNP is situated in the extreme northern part of the Northern Cape Province in South Africa and forms part of the |Ai-|Ais/Richtersveld Transfrontier Park. The RNP is unusual in that human inhabitants and alluvial diamond mining operations have continued to occupy the park since its proclamation in 1991. The South African National Parks (SANParks) manages the park as a communal pasture on a contractual basis, and 26 registered farmers from the local Nama community have the right to graze the equivalent of a total of 6,600 head of small livestock units (SLU) within the park boundaries.

The initial 30-year contract park agreement that came into effect in 1991 will expire in 2021, but a new agreement has been negotiated and is expected to be signed in the near future, thereby ensuring continued protection of the park for at least another 30 years. The extreme aridity of the environment means that alternative uses are unlikely to be viable and the park is thus likely to persist in the long-term.

The climate of the RNP is harsh; mean annual precipitation (MAP) ranges from ca. 55 mm in the north and northeast to ca. 125 mm in the southwest of the park (WorldClim dataset, Hijmans et al. 2005), with peak temperatures at times exceeding 50 °C (SANParks 2018). In addition, rainfall is highly sporadic and patchy and some localities may receive no rain for several years; a site in the Tatasberg had no rainfall recorded from 2000–2016, during which period the average MAP for the park interior was 38 mm and one site received 163 mm MAP (SANParks 2018). While species endemic to the area clearly must be adapted to such conditions, changing climate may result in some species’ tolerances being exceeded.

From 1990 to 2010 the RNP had already experienced an increase of 1.2 °C in mean maximum and 1.1 °C in mean minimum temperatures, exceeding the near-future (2035) predictions utilised by the South Africa Department of Environmental Affairs (DEA) for planning purposes, and there had been a significant increase in the number of days exceeding 35 °C (Wilgen et al. 2015). In conjunction with no change in MAP (but with a higher number of smaller rainfall events) this indicates an increase in aridity in the region from 1990–2010. A subsequent severe drought from 2016 to present, during which Sendelingsdrift (23 km west of the type locality of T. nama sp. nov.) has received an annual average of just over 6% of its historic MAP for four years, has resulted in large-scale vegetation loss. Even extreme arid-adapted species such as the three quiver tree species (Aloidendron dichotomum (Masson) Klopper & Gideon F. Sm., A. pillansii (L. Guthrie) Klopper & Gideon F. Sm. and A. ramosissimum (Pillans) Klopper & Gideon F. Sm.) present in the park have experienced extreme die-off, with an approximately 70% reduction in numbers of live trees (B. Whittington, pers. comm.). If such changes continue and the further temperature rises predicted for the coming decades occur, the RNP will become increasingly desert-like (SANParks 2018).

It is thus likely that local conditions could become unsuitable for species such as Tetramorium nama sp. nov. and they would have to track suitable habitat and climatic conditions in order to survive. Conditions suitable for nuptial flights often would not occur for periods of several to many years in the northern RNP and it is only by flighted dispersal that these ants could migrate far enough to track substantial climatic changes. Given the rapidity of recent climatic change it is possible that they might not be able to respond quickly enough and might therefore become locally or even globally extinct.

Alluvial diamond mining has been carried out along the banks of the Orange River in the Richtersveld since around 1900; mining at Oena started in 1992, though the mine has changed ownership since then and is currently operated by African Star Minerals (ASM). Mining operations have been carried out at three areas (Oena, Sandberg and Blokwerf) within the mineral lease area, but have at least temporarily been suspended at one (Blokwerf). Mining operations are opencast, with topsoil (where present) being stripped and stockpiled, after which the alluvial gravels are excavated, screened and processed through rotary pans to provide concentrate for extraction in the recovery plant. Approximately 80% of the coarse waste is returned to the excavated areas to be used as backfill, with the remainder deposited in waste rock dumps. The fine tailings slurry is deposited in a dam and will eventually either be covered with coarse tailings or returned to excavated areas as backfill. The backfilled pits are covered with topsoil where this is available (topsoil is absent from much of the terrace gravel areas) and the area re-contoured to a natural-looking state. However, virtually all plants and animals within the mined deposits and in areas where the processed gravels are dumped will be killed. The resulting soil structure will also probably differ very substantially from the natural alluvial deposits and rehabilitation can be expected to be extremely slow given the arid nature of the environment. All mined areas and those covered by waste rock or tailings are therefore effectively lost as habitat for indigenous species for a considerable period, which may be many decades. The area likely to be disturbed by currently planned mining activities constitutes only ca. 20% of the alluvial plains within the Oena mineral lease area, but this could be expanded (prospecting is continuing at two sections upstream of the three areas listed above) and a significant area had been disturbed by previous operations and has not yet been rehabilitated; this is to be rectified as part of the current environmental management plan (ASM 2018). The total area of habitat loss could therefore be a significant proportion of the total available.

The potential impact of artificial lighting of mining operations on Tetramorium nama sp. nov. cannot be determined at this stage as it is not known whether the nuptial flights of this species take place during the day or night. Impacts on nocturnal nuptial swarms could be severe, especially considering the extremely sporadic nature of suitable conditions for such events. Dust is another potentially significant factor that may reduce food availability via impacts on plants.

Members of the local Nama community run herds primarily of domestic goats, Capra aegagrus hircus (Linnaeus, 1758), with the total domesticated livestock population within the RNP capped at 6,600 small livestock units (a stocking rate of approximately 25 ha per SLU). However, the riverine fringe along the banks of the Orange River where Tetramorium nama sp. nov. was collected is probably the most intensively utilised habitat within the park, as it provides fodder during the long dry periods when the vegetation further from the river provides very little forage (SANParks 2018).

Degradation of vegetation by goat overgrazing is recognised to be more severe than that from any other ruminant livestock species, due to their ability to graze on residual biomass and woody species that would be left as vegetation cover by other livestock (Steinveld et al. 2006), as well as their ability to debark trees, range over large distances and survive without water for longer periods than other livestock (Lipson et al. 2011).

Intensive grazing even by domesticated goats would be expected to substantially reduce seed set by grasses and shrubs that provide food for Tetramorium solidum-group species. Continuation of the current trend of climate change is likely to reduce carrying capacity in the RNP and lead to an increased risk of overgrazing; regular review of grazing quotas should ideally be carried out as part of the park’s Environmental Management Plan. Should feral populations of goats become established in the park, the cap on numbers would become very difficult to maintain and impacts could become far more severe, potentially virtually eliminating the food supply of granivorous ants and putting them at high risk of at least local extinction.

However, T. signatum Emery, 1895, another T. solidum-group species, has been observed (unpublished data) preying on live Microhodotermes viator (Latreille, 1804), which suggests that members of the group might not be wholly dependent on seeds as a food source. The extent to which termites and perhaps other insect prey can substitute for their normal food is unknown, but this may at least partially mitigate the impact of livestock such as goats on their food supply; further research is required.

Distribution data are, at present, too limited to allow a full IUCN Red List assessment (IUCN 2012) to be carried out and T. nama sp. nov. should currently be considered Data Deficient (DD). However, it is very likely that the species will prove to be restricted to the northernmost parts of the Northern Cape and southernmost parts of Namibia, given the lack of any records of the species from numerous sampling events to the north and south of this region. In light of the multiple threats discussed above (each of which is expected to contribute to continuing decline in quality and extent of habitat), even if the range proves to be as much as four times that of the entire RNP, the Extent of Occurrence (EOO) would fall within the range (< 20,000 km²) for the species to be classified as Vulnerable (VU) under criterion B1ab(iii). The species may however be limited to a narrower region straddling the Orange River, which would lead to an Endangered (EN) classification under the same criterion if the EOO proves to be less than 5,000 km². Other ant species might be similarly threatened by the factors discussed here, although the impact of grazing would be expected to vary depending on their reliance on seeds or other plant-related food sources (such as honeydew from phytophagous homopterans).