Lemniscomys rosalia – Single-striped Grass Mouse Assessment Rationale Listed as Least Concern in view of its wide distribution within the assessment region (occupying an array of grassland habitats), its occurrence in many protected areas (including Kruger National Park), its potential range expansion in North West Province, its tolerance of habitat modification, and because there are no major threats that could cause population decline. Research has shown that its diet consists of the most abundant grass species occurring in a particular habitat, not necessarily the most nutritious, and thus food resources may not be a limiting factor in its distribution. However, land managers should Angelique Lazarus employ ecological stocking levels to avoid overgrazing and thus sustain tall grasses. Prudent fire management is recommended too. Regional Red List status (2016) Least Concern Regional population effects: As suitable habitats are National Red List status (2004) Data Deficient suspected to be largely connected along the northern Reasons for change Non-genuine: border of South Africa with Botswana, and through change in risk Mozambique and Zimbabwe through the Great Limpopo tolerance Transfrontier Conservation Area, as well as between Mozambique and northern KwaZulu-Natal Province, Global Red List status (2016) Least Concern rescue effects may be possible. TOPS listing (NEMBA) (2007) None CITES listing None Distribution Endemic No This species is widely distributed across the savannah regions of Namibia, South Africa, Swaziland, Zimbabwe, This species feeds on the most abundant central and northern Botswana, Angola, Mozambique, grass species of a habitat, and thus is not reliant on Zambia, Malawi, Tanzania and southern Kenya high-quality grasses, and it is found both in (Monadjem et al. 2015). Within the assessment region, it experimentally herbivore-dominated and herbivore- occurs in the Savannah Biome in landscapes with tall, excluded plots (Hagenah et al. 2009). dense grass, including agricultural landscapes, distributed across KwaZulu-Natal, Gauteng, Mpumalanga, North West and Limpopo provinces. Power (2014) presents evidence for a possible range expansion in North West Province: aside from being Taxonomy recorded throughout the mesic savannah regions, it was Lemniscomys rosalia (Thomas 1904) captured on one particular site on Khamab Kalahari Reserve in the Molopo Plains Sandy Bushveld vegetation ANIMALIA - CHORDATA - MAMMALIA - RODENTIA - type, which expands the range by 250 km and represents MURIDAE - Lemniscomys - rosalia South Africa’s westernmost record (Figure 1). This may be Common names: Single-striped Grass Mouse, Zebra a genuine range expansion linking to Botswana’s Kalahari Mouse (English), Eenstreepmuis (Afrikaans) population (Smithers 1971), due to the high intensity of small mammal surveying in the Molopo region (Newbery Taxonomic status: Species 1996). Taxonomic notes: Meester et al. (1986) listed four subspecies of which only one, Lemniscomys r. spinalis Population (Thomas 1916) occurs in the assessment region: KwaZulu- Natal, Mpumalanga, Limpopo, Gauteng and North West The Single-striped Grass Mouse is relatively common to provinces, as well as Swaziland and western Zimbabwe. very common across its range. For example, it was the Van der Straeten et al. (2008) note that L. rosalia may most frequently encountered species in Hluhluwe-iMfolozi represent a complex of species. However, as only one Park, KwaZulu-Natal Province, representing 75% of all subspecies is listed for the assessment region, and due to small mammal samples (Hagenah et al. 2009). In contrast, the lack of either morphometric or genetic data, we treat it it was one of the least abundant murids captured at as a single species for this assessment. uMkuze Game Reserve, Kube Yini Game Reserve and Phinda Private Game Reserve, KwaZulu-Natal (Delcros et al. 2015; Rautenbach et al. 2014). It occurs in high densities in suitable habitat. For example, in an ungrazed Recommended citation: Monadjem A, Schoeman C, Child MF. 2016. A conservation assessment of Lemniscomys rosalia. In Child MF, Roxburgh L, Do Linh San E, Raimondo D, Davies-Mostert HT, editors. The Red List of Mammals of South Africa, Swaziland and Lesotho. South African National Biodiversity Institute and Endangered Wildlife Trust, South Africa. The Red List of Mammals of South Africa, Lesotho and Swaziland Lemniscomys rosalia | 1 Figure 1. Distribution records for Single-striped Grass Mouse (Lemniscomys rosalia) within the assessment region Table 1. Countries of occurrence within southern Africa Habitats and Ecology Country Presence Origin This species tolerates a wide range of savannah habitats, Botswana Extant Native including savannah woodland, dry grassland at the ecotone between vleis and woodland (for example, Fuller Lesotho Absent - & Perrin 2001), and scrub savannah of the Kalahari, but Mozambique Extant Native the common factor to all habitats is tall grassland areas Namibia Extant Native (Kern 1981; Bowland & Perrin 1988; Monadjem 1997a; Skinner & Chimimba 2005; van der Straeten et al. 2008; South Africa Extant Native Rautenbach et al. 2014). It was sampled in tall closed Swaziland Extant Native grassland, low open rocky woodland, cultivated areas and open woodland in the Maguga Dam area, Swaziland Zimbabwe Extant Native (Avenant & Kuyler 2002). It is known to inhabit the tall grasses surrounding agricultural lands, especially fallow fields, and also occupies suburban gardens in KwaZulu- subtropical grassland near Matsapha, Swaziland, Natal (Taylor 1998; Skinner & Chimimba 2005). It prefers densities ranged from 4.4 ± 0.8 to 8.2 ± 1.0 individuals / areas where there is a good cover of grass, herbs or reeds ha from 1995 to 1998 (Monadjem & Perrin 2003). Although and avoids areas where vegetation cover is short or its abundance fluctuates widely, it is more numerous in absent (Skinner & Chimimba 2005). It vacates recently dry winter months (Monadjem & Perrin 2003). burnt areas, but returns as soon as the grass recovers Current population trend: Stable, inferred from stable (Monadjem & Perrin 1997). It has not been recorded from savannah habitat. montane grasslands, forests and riverine woodlands (Monadjem 1999; Delcros et al. 2015). Continuing decline in mature individuals: No It is predominantly crepuscular but can be diurnal (Perrin Number of mature individuals in population: Unknown 1981) and terrestrial; burrows are excavated where there Number of mature individuals in largest subpopulation: is overhead cover of matted grass and form runs from Unknown burrow entrances to foraging areas (Skinner & Chimimba 2005). In a study in Kruger National Park, stomach content Number of subpopulations: Unknown analysis revealed a diet comprised of 50% seeds, 48% herbaceous material and 2% insects (Watson 1987). Severely fragmented: No. Can utilise transformed and Similarly, in Hluhluwe-iMfolozi Park, KwaZulu-Natal, grass agricultural landscapes. leaves and stems took up 65% of the diet while seeds (25%) and arthropods (3%) comprised the bulk of the Lemniscomys rosalia | 2 The Red List of Mammals of South Africa, Lesotho and Swaziland Table 2. Threats to the Single-striped Grass Mouse (Lemniscomys rosalia) ranked in order of severity with corresponding evidence (based on IUCN threat categories, with regional context) Evidence in the Scale of Rank Threat description Data quality Current trend scientific literature study 1 2.3.2 Livestock Farming & Bowland & Perrin 1989 Empirical Local Possibly increasing with human Ranching: reduction of settlement expansion and intensification grassland height. of wildlife farming. 2 7.1.1 Increase in Fire Kern 1981 Empirical Local L. rosalia was not recorded in burn plots. Frequency/Intensity: reduction of grassland height. Bowland & Perrin 1988 Empirical Local L. rosalia survival higher in non-burn plots. remaining food items (Hagenah et al. 2009). In Swaziland, overgrazing that reduces tall grass swards (Bowland & herbaceous material comprised 84.6 ± 9.2% of the diet, Perrin 1989), and overly frequent fire regimes that have and seeds comprised 15.4 ± 9.2% (Monadjem 1997b). the same effect (although it may return after grass has Hagenah et al. (2009) demonstrated that it prefers the recovered, Monadjem & Perrin 1997). Such management most abundant grass species occurring in its habitat and practices will decrease habitat quality and may reduce the not necessarily the highest quality species, and suggested area of occupancy. that the arthropod component of the diet may contribute Current habitat trend: Savannah ecosystems are not significantly to meet the nutritional requirements of the threatened (Driver et al. 2012), and this species is tolerant species. In Hluhluwe, it mostly fed on two low-quality of modified landscapes and agricultural areas. bunch grass species, Sporobolus africanus (69%), and Eragrostis curvula (21%), whereas, in iMfolozi, it predominantly fed on the high-quality lawn grass species Conservation Urochloa mosambicensis (77%) and Themeda triandra, a lower-quality bunch grass species, comprised the This species is present in many protected areas within the remainder of its diet (13%). Grass species composition is assessment region, including Kruger National Park, therefore unlikely to be a limiting factor in the diet of this uMkuze Game Reserve and Phinda Private Game species. Its most important habitat requirement is Reserve. No interventions are necessary at present. The probably dense ground cover of long grass as it breeds in expansion