Mitt. Mus. Nat.kd. Berl., Zool. Reihe 80 (2004) 2, 261-273 05.09. 2004 Ants along a southern African transect - a basis for biodiversity change monitoring (Insecta, Hymenoptera, Formicidae)' Frank Koch2 & Katrin Vohland2 With 10 figures and 1 table Abstract Up to now, 24 ant species have been collected during field studies along a southern African transect in Namibia and South Africa as part of the BIOTA project - Biodiversity Monitoring Transect Analysis in Africa. Some of these, Ocyrnyrrnex hirsutus, 0. resekhes, Tetrurnoriurn erecturn, Meranoplus glaber, and M. rnagrettii were recorded for the first time in Namibia. For other species the occurrence in specific habitat types was confirmed. A summary of relevant morphological and ecological characteristics of each species is given. For each genus an example is illustrated. This study assesses the value of ants for biodiversity monitoring in southern Africa. Ants are particularly suited for monitor- ing approaches, because of their high abundance and species diversity, regular occurrence throughout the year as well as nest stability. Problems may arise from the unconsolidated biosystematic status in some genera, namely Tupinorna and Puchycon- dylu, and the lack of detailed information for many species on their habitat requirements. Key words: Formicidae, Namibia, South Africa, biodiversity, ecology, trophic guilds, distribution. Introduction This study presents results gained as part of the Biodiversity Transect Analysis in Africa (BIOTA). BIOTA aims to monitor biodiversity long term under environmental and climate change. In the first pilot study phase, a base- line of organisms suitable as change indicators in the sub-Saharan ecosystems had to be iden- tified. Ants are rather well studied organisms. For many of them detailed ecological information is available. Ants have a large impact on ecosys- tems simply due their biomass. They alter nutri- ent cycling pathways and the spatial distribution of nutrients, they impact on soil structure and can act as seed dispersers (Holldobler & Wilson 1990, Elmes 1991, Majer 1993, Agosti et al. Ants have a number of characteristics which make them suitable as indicators of environmen- tal change (Alonso 2000): ants are abundant in most ecosystems, their occurrence is strongly cor- related with biotic features, they are present 2000). throughout the year. However, difficulties may arise when differentiating species composition, e.g. comparing ant diversity between differently used areas and road margins (Tshiguvoho et al. 1999). The most abundant and ecologically important species occurring along the BIOTA transect are presented here, and their distribution is dis- cussed in relation to the trapping design. Analy- sis also considers the trophic guild of the ants. According to Brown (2000) these can be differ- entiated into predators (P) which actively hunt their prey; scavengers (S) which belong to gener- alistic feeders, consuming dead arthropods, litter and seeds; honeydew feeders (H) living in close association with phloem sucking insects like aphids or coccids; granivorous feeders (G) which harvest seeds, often from grasses, and a mixture of different feeding types. This article is also meant to encourage stu- dents and scientists of biodiversity monitoring and ecosystem studies by providing habitus sketches which allows the identification of gen- era in the field. Dedicated to Prof. Dr. Michael Stubbe for his 65'h birthday. Institut fur Systematische Zoologie, Museum fur Naturknnde, Humboldt Universitat, Invalidenstr. 43, D-10115 Berlin. E-mail: [email protected]Received March 2004, accepted April 2004
Ants along a southern African transect - a basis for biodiversity change monitoring (Insecta, Hymenoptera, Formicidae)'
Frank Koch2 & Katrin Vohland2
With 10 figures and 1 table
Abstract
Up to now, 24 ant species have been collected during field studies along a southern African transect in Namibia and South Africa as part of the BIOTA project - Biodiversity Monitoring Transect Analysis in Africa. Some of these, Ocyrnyrrnex hirsutus, 0. resekhes, Tetrurnoriurn erecturn, Meranoplus glaber, and M. rnagrettii were recorded for the first time in Namibia. For other species the occurrence in specific habitat types was confirmed.
A summary of relevant morphological and ecological characteristics of each species is given. For each genus an example is illustrated.
This study assesses the value of ants for biodiversity monitoring in southern Africa. Ants are particularly suited for monitor- ing approaches, because of their high abundance and species diversity, regular occurrence throughout the year as well as nest stability. Problems may arise from the unconsolidated biosystematic status in some genera, namely Tupinorna and Puchycon- dylu, and the lack of detailed information for many species on their habitat requirements.
This study presents results gained as part of the Biodiversity Transect Analysis in Africa (BIOTA). BIOTA aims to monitor biodiversity long term under environmental and climate change. In the first pilot study phase, a base- line of organisms suitable as change indicators in the sub-Saharan ecosystems had to be iden- tified.
Ants are rather well studied organisms. For many of them detailed ecological information is available. Ants have a large impact on ecosys- tems simply due their biomass. They alter nutri- ent cycling pathways and the spatial distribution of nutrients, they impact on soil structure and can act as seed dispersers (Holldobler & Wilson 1990, Elmes 1991, Majer 1993, Agosti et al.
Ants have a number of characteristics which make them suitable as indicators of environmen- tal change (Alonso 2000): ants are abundant in most ecosystems, their occurrence is strongly cor- related with biotic features, they are present
2000).
throughout the year. However, difficulties may arise when differentiating species composition, e.g. comparing ant diversity between differently used areas and road margins (Tshiguvoho et al. 1999).
The most abundant and ecologically important species occurring along the BIOTA transect are presented here, and their distribution is dis- cussed in relation to the trapping design. Analy- sis also considers the trophic guild of the ants. According to Brown (2000) these can be differ- entiated into predators (P) which actively hunt their prey; scavengers (S) which belong to gener- alistic feeders, consuming dead arthropods, litter and seeds; honeydew feeders (H) living in close association with phloem sucking insects like aphids or coccids; granivorous feeders (G) which harvest seeds, often from grasses, and a mixture of different feeding types.
This article is also meant to encourage stu- dents and scientists of biodiversity monitoring and ecosystem studies by providing habitus sketches which allows the identification of gen- era in the field.
Dedicated to Prof. Dr. Michael Stubbe for his 65'h birthday. Institut fur Systematische Zoologie, Museum fur Naturknnde, Humboldt Universitat, Invalidenstr. 43, D-10115 Berlin.
Koch, F & K. Vohland, Ants along a southern African transect
Mile 46 Research Station
MNHU = Museum fur Naturkunde, Humboldt- Universitat zu Berlin.
TL = Total Lenght (TL) of workers: The fully extended length of the indivi- dual, from mandibular apex to gas- tral apex
Material and methods
Study Area
The study area consisted of a transect of plots, called “biodi- versity observatories” which were sampled in close collabora- tion with other subprojects of BIOTA. The transect ranges from Mile 46 Research Station in northern Namibia to Quag- gafontein in Namaqualand, Northern Cape Province, South Africa (Fig. 1, right). Within the the zoological subproject of the Museum fur Naturkunde, Berlin, ten biodiversity obser- vatories are sampled with variable intensity (Fig. 1, left).
The plot Mile 46 Research Station is situated about 100 km south of Rundu, Namibia. It belongs to old Kalahari sand dune systems, and consequently the soils are classified as Arenosols (Petersen, pers. comm.). The vegetation type is Kalahari woodland, with Guibourtia coleosperma (Benth.) J. LConhard (Caesalpinioideae) as dominant tree species in the plots and Melinis repens (Willd.) Zizka (Poaceae) as most frequent grass species. It is a governmental research station for cattle breeding and is used with low intensity below car- rying capacity, but is impacted by frequent fires.
Toggekry
Toggekry is situated about 100 km north of Okahandja, Na- mibia. The plots are characterized by Arenosols and Cambi- sols (Petersen, pers. comm.). The tree savanna is dominated by the acacia species Acacia mellifer (Vahl) Benth, A. tortilis (Forsk.) Hayne, A. reficiens Wawra (Mimosoideae) and Bos- cia albitrunca (Burch.) Gilg & Benedict (Capparaceae); main grass was Sfipagrostis uniplumis (Licht.) De Winter (Poa- ceae).
The plot belongs to a privately owned farm used commer- cially for game hunting.
0 75 150 300 450 60R,,ometers I=-
Angola I Zambia (
) Namibia 9 < Botswana)
iJ-+c outh Africa
Legend
Biodiversity Observatories
Subtropical woodland
Tree and shriib savanna
Nama Karoo a Succulent Karoo
a Namib Desert
0 Frontiers
Lakes and salt pans
N
Fig. 1. Map showing the investigated biodiversity observatories in Namibia and South Africa. Data modified (after Mendel- sohn et al. 2002).
Mitt. Mus. Nat.kd. Berl., Zool. Reihe 80 (2004) 2 263
Otjiamongombe (Erichsfelde)
Otjiamongombe is situated about 20 km from Toggekry. A small rivier (ephemeral river, river flowing after heavy rains) crosses the plot, The soils in the plots were classified as Cam- bisols and Luvisols (Petersen, pers. comm.). It has a more diverse surface structure than Toggekry (Horsch, pers. comm.), and a high abundance of the shrub Monechma geni- stifolia (Engl.) c. B. Clarke (Acanthaceae). In contrast to Toggekry, Otjiamongombe is intensely used for cattle.
Gellap-Ost
Gellap-Ost is situated in the dwarf shrub savanna of the Nama Karoo, about 20 km west of Keetmanshoop, Namibia. The soil of the plots is classified as eutric Leptosol, Cambi- sol, and Regosol, respectively (Petersen, pers. comm.), and the plot structured by volcanic outcrops. During the whole study period, the surface was covered by Stipagrostis uniplu- mis (Licht.) De Winter (Poaceae). Dominant bushes are Rhi- gozum trichotomum Burch. (Bignoniaceae) and Catophractes alexandri D. Don (Bignoniaceae). The plot belongs to a gov- ernmental research station for breeding Karakul sheep.
Nabaos
The Nabaos plot is bordering to the Gellap-Ost plot, only divided by a fence. The dominant soils are classified as Rego- sols (Petersen, pers. comm.). Most frequent shrubs are Rhi- gozum trichotomum and Calicorema capitata (Moq.) (Ama- ranthaceae). In contrast to Gellap-Ost, it is communally used for keeping goat and sheep. It is strongly degraded and most of the time bare of ground vegetation cover (Hoffmann & Zeller 2003, Vohland et al. 2004).
Karios
Karios is part of privately owned land close to the Fish River Canyon Park, Namibia. It bears volcanic outcrops in Regosol (Petersen, pers. comm.). The plot lies within the Nama Ka- roo but, in contrast to Gellap-Ost and Nabaos, Karios is domi- nated by the Euphorbia gregaria Marloth (Euphorbiaceae).
Numees and Koerogabvlakte (Richtersveld)
Numees as well as the Koerogabvlakte are situated within the Richtersveld National Park, in northern Northern Cape Province, South Africa. The soils are highly variable, while Numees is dominated by Regosols, in the plains of the Koero- gabvlakte Calcisols, Cambisols as well as Durisols are found (Petersen, pers. comm.). This region receives winter rain and belongs to the Succulent Karoo. Numees is situated at lower altitude than the Koerogabvlakte. Dominant plants are succu- lents as e.g. Brownanthus pseudoschlichtianus Pierce $2 Ger- baulet.
Both areas are used for keeping goats, but in the Koero- gabvlakte more damage and tunnel erosion is visible.
Groot Derm (Yellow Dunes)
Groot Derm is situated in the northernmost parts of the Suc- culent Karoo, in the sand dunes about 30 km NE to Alex- eander Bay, Northern Cape Provine, South Africa.
Quaggafontein
Quaggafontein is situated in Namaqualand, Northern Cape Province, South Africa. Soil type is described as Leptosol (Petersen, pers. comm.). The soil is strongly modified by ter- mites, which is expressed by “heuweltjies” (eroding termite mounds). The plot is dominated by Asteraceae. It is com- munally used for sheep and goat farming, and lies at the foot of a small hill.
Sampling methods
Sampling took place once a year, at end of the rainy season. Subsequently in 2001, 2002 and 2003 samples were taken in March and April in the summer rain plots of Namibia, which are Mile 46 Research Station, Toggekry, Otjiamongombe, Gellap-Ost, Nabaos and Karios, and in October in the winter rain plots of South Africa, namely Numees and Koerogabv- lakte in the Richtersveld and Quaggafontein in Namaqua- land. Groot Derm was only visited once because it did not belong to the priorized plots.
Sampling was strictly standardized by ten pitfall traps with a diameter of 87 mm which were set in a line on those two to three subplots at each biodiversity observatory which were ranked as priority plots in the frame of the BIOTA project. The traps were filled with monoethylen glycol, and were al- ways set for four days.
The animals were stored in a solution of 60% alcohol, 30% purified water and 10% acetic acid and transfered to the laboratory. The material is shared with the National Mu- seum of Namibia, Windhoek, and the Plant Protection Re- search Institute, Pretoria, South Africa, respectively.
The taxonomical revision of the Camponotus maculatus (Fabricius)-group is in preparation and therefore this materi- al is completely deposited in MNHU.
Results and discussion
Methodological considerations
Pitfall trapping is a common and widely applied method in ant studies (e.g. Lindsey & Skinner 2001; Koivula et al. 2003). In this study, numbers of individuals were not considered as they are known to vary depending on the actual distance to the nests, resource availability and climate. To study specifically community structure of ants a more elaborated study design would have been necessary (e.g. Agosto et al. 2000), whereas in this study pitfall traps were in first instance used to investigate on potential dietary resources for small mammals.
For monitoring of environmental changes, long term studies over three years or more would be necessary as populations fluctuate interannually (Kaspari & Majer 2000). In this case, Groot Derm has only be visited once, while data from Otjiamongombe, Gellap-Ost, Nabaos, Karios, and the Koerogabvlakte circumscribed three vis- its in three years, and gave reliable results.
The ants, which have been collected in the pit- fall traps during the three years of the pilot phase can be confined to four subfamilies, nine genera, and 24 species (Tab. 1). In general, the further north, the more species were recorded, ranging from 13 species in Mile 46 Research Sta- tion to only four species in the Richtersveld. In Groot Derm only two species were recorded, but here sampling was much shorter than anywhere else.
Mitt. Mus. Nat.kd. Berl., Zool. Reihe 80 (2004) 2 265
A comparison of newly detected with earlier re- corded species yielded some surprises. Not all ex- pected species were found. That probably has its reasons in the BIOTA approach which does not aim to inventory all species present in a plot, but to set a baseline to monitor changes in typical habi- tats. One shortcoming therefore was that the plots were determined using a computer programme to fullfill standardization requirements. Therefore the plots were selected quasi a priori instead of firstly survey the area and then identify typical habitat and microhabitat structures for plot identification. At present, more detailled information from the other subprojects exist which might improve plot selection for ant monitoring in future.
Subfamily Formicinae
Camponotus Mayr, 1861
The genus Camponotus is distributed world-wide and belongs to the largest and most diverse groups of ants. About 935 species are described. They are well distributed throughout the Afro- tropical region and more than 70 species have been recorded from southern Africa. The taxon- omy of the Afrotropical species is confuse and identifications are uncertain, thus the literature on distributional records is unreliable. Smaller revisions of species groups exist from Baroni Ur- bani (1972) for the C. maculatus (Fabricius)- group and from Robertson & Zachariades (1997) for the C. fulvopilosus (De Geer)-group.
The nests are variable in size and are built subterraneous, in termite mounds, in dead wood in and on trees (Brown 2000).
Camponom-species are generalistic foragers. They also feed on honeydew from Aphidina or Coccina and sometimes also prey on termites.
The genus contains medium-sized to large, polymorphic ants. Therefore, the total length of workers is very variable.
In this study, four Camponotus-species were con- sidered. Camponotus detritus Emery, 1886 seems to be endemic in the dune-fields of the Namib Desert (Marsh 1986a, Robertson & Zachariades 1997). Therefore the occurrence of this species in the southwestern part of the transect is expected.
Camponotus exsanguis Forel, 1910
TL: 9.5-15.0 mm. This species was described by Forel (1910), and based on the the very light reddish-yellow vari-
ety C. mystaceus var. exsanguis from C. mysta- ceus Emery with the locus typicus: “Prince of Wales Bay, siidlich von Liideritzbucht”. There- fore the data on distribution and ecology of this species has to be confirmed (Marsh 1986a, b), as the identity of C. exsanguis and/or C. mystaceus is not clear-cut. Probably Marsh (1986a, b) con- sidered C. exsanguis as a synonym of C. mysta- ceus. This view is supported by comments on the couloration of the ants from the central Namib Desert “translucent; pale brown but sometimes with dark brown head and gaster” (Marsh 1986a).
In order to gain reliable data on the distribu- tion of C. exsanguis, all material of the MNHU was examined. According to this investigated specimens it seems that this species is only dis- tributed in South Africa and Namibia. In this context it is necessary to collect more material and to examine additionally collections. The nests were built subterraneous in sandy soils (Forel 1910).
Head, mesosoma and gaster yellowish; poster- ior half of clypeus with a distinct raised median step with a transverse row of strikely long and stout, curved hairs, a line of such hairs present on dorsal and ventral surface of the mandibles, in ventral view similar hairs present on the basic of mentum of the mouth parts.
Camponotus fulvopilosus (De Geer, 1778)
TL: 10.0-16.0 mm; Fig. 2. Camponotus fulvopilosus was intensively treated in the revision of the C. fulvopilosus-group (Ro- bertson & Zachariades 1997). The authors distin- guished between five morphotypes depending on the habitat type ranging from the Namib Desert to the savanna. In this study, only the savanna form was collected.
This species is distributed throughout most of the arid savanna and woodland ecosystems of southern Africa and its distribution extends as far north as Democratic Republic of Con- go.
Camponotus filvopilosus is mainly diurnal and collects honeydew from Aphidina or Coccina (Sternorrhyncha) which are feeding on Acacia- trees (Marsh 1986a).
Head and mesosoma black or brownish-black, gaster black; dorsal surface of the gaster with pale thick hairs and more or less glabrous patches medially.
Camponotus maculatus (Fabricius, 1782) Camponotus mystaceus Emery, 1886
T L 7.7-12.0 mm. This species is distributed in northern Africa as well as in the Afrotropical region and the Mada- gascan subregion. In southern Africa, C. macula- tus belongs, next to C. mystaceus and C. fulvopi- losus, as the most common ant species. Accordingly, a wide range of varieties and sub- species were described on C. maculatus (Mayr 1862, Baroni Urbani 1972, Bolton 1995). In Na- mibia this species is widely distributed on the gravel plains of the central Namib Desert.
Members of this species are nocturnal and pre- dominantly honeydew feeders.
Head, mesosoma and gaster yellow to light brown; clypeus and mandibles without strikingly long and stout, curved hairs (psammophore).
TL: 7.0-15.0 mm. According to Arnold (1924) this species is dis- tributed in southern Africa. In the central Namib Desert it occurs across the desert corning close to the coast (Marsh 1986b). Also here data on distribution and ecology (Marsh 1986a, b) has to be checked.
The members of this species are nocturnal (?) and predominantly honeydew-feeders from Aphidina or Coccina occuring on Stipagrostis Nees spp. (Poaceae). Marsh (1986a) also ob- served, that workers took nectar from Arthrue- m a leubnitziae (Kuntze) Schinz flowers (Ama- r anthaceae) .
Head reddish-brown, dorsally dark-brown, me- sosoma reddish-brown and gaster dark brown.
Mitt. Mus. Nat.kd. Berl., Zool. Reihe 80 (2004) 2 267
The long curved hairs on the posterior half of clypeus, on mandibles and mentum similar to C. exsanguis.
group, the 0. weitzeckeri-group and the 0. barbi- ger-group.
Ocymyrmex hirsutus-group Subfamily Myrmicinae
Ocymyrmex Emery, 1836
Species assigned to the genus Ocymyrmex are confined to the Afrotropical region. In rain for- ests and humid savannas of the Guineo-Congo- lian region no species were recorded so far, probably because Ocymyrmex-species prefer arid regions. The situation in the Sahelian subregion south of the Sahara is different from Eastern Africa where only few species were recorded. Most species have been described in southern Africa, with Namibia and South Africa being most specise (Bolton 1981a, Bolton & Marsh 1989).
From the 37 valid species, 17 species were documented from Namibia and 17 from South Africa (Bolton & Marsh 1989). However, no val- id information exists for the real distribution of single species, as also shown in this study.
There are two important reasons for the lack of knowledge. First the sparse collections, and second the unconsolidated taxonomic knowl- edge. After the first revision of this genus from Bolton (1981a), from 25 described species and varieties 16 remained valid, while 7 were newly confined to this genus. After a further revision from Bolton & Marsh (1989) the number in- creased to 37 species, from which 9 from south- ern Africa were newly described. From these, 6 species have been reported from Namibia, one species from Namibia and South Africa, one from South Africa and one from Botswana and South Africa.
Ocymyrmex-species build their nests below the ground in sandy or stony soils. Next to Ocymyr- mex, only the vicariant taxon Cataglyphis Forster from northern Africa and Eurasia live in com- parable habitats. They are also diurnally active and extremely thermophilous. This highly specia- lized adaptation allows these species to forage during the hottest parts of the day (Wehner et al. 1983, Holldobler & Wilson 1990).
In general, Ocymyrmex-species are seed har- vesters (Brown 2000), but some also predate on termites with different intensity (Marsh 1986a, Bolton & Marsh 1989).
With this study, only those species were re- corded that belong to the Ocymyrmex hirsutus-
Ocymymzex hirsutus Forel, 1910
TL: 6.9 - 7.8 ~llfn.
This species only was known from the locus typi- cus in Botswana “Sevrelala, Kalahari“ and South Africa. This study provided the first record for Namibia.
The single locality “Mile 46 Research Station” is characterized by the vegetation type “North- eastern Kalahari Woodland’. The locus typicus also is part of the tree and shrub savanna. The nests are subterraneous (Forel 1910).
Colour of head, mesosoma and gaster reddish; sometimes the mesosoma and gaster slightly dark- er.
Ocymyrmex resekhes Bolton & Marsh, 1989
TL: 7.9-8.6 mm; Fig. 3. This species was previously only known from the northern Cape Province, South Africa. Accord- ing to the collection sites in the southern Kala- hari National Park, Bolton & Marsh (1989) as- sumed that 0. resekhes might be distributed within Namibia and Botswana, too. The record of 0. resekhes at Mile 46 Research Station in northern Namibia confirms the occurrence of this species in Namibia.
Colour of head and mesosoma reddish-brown, gaster lighter.
Ocymyrmex weitzeckeri-group
Ocymyrmex fortior Santschi, 1911
TL: 6.7-8.2 111111. Ocymyrmex fortior is the most widely distributed species in southern Africa. It was recorded in Angola, Botswana, Namibia, South Africa, Zam- bia and Zimbabwe (Bolton & Marsh 1989). Jud- ging from known habitat requirements and local- ities ranging from the Kruger National Park in the province Mpumalanga and KwaZulu-Natal to the biodiversity observatories in northern Nami- bia, this species shows mesotopic characteristics.
Ocymyrmex fortior is mainly granivorous but hunting and capturing of workers of Odontoma-
268 Koch, F & K. Vohland, Ants along a southern African transect
chus troglodytes (Ponerinae) and preying on ter- mites was also found (Bolton & Marsh 1989). In Mile 46 Research Station 0. troglodytes lives in the same habitat as its predator.
Colour of head and mesosoma brick-red to lighter red, gaster the same colour or slightly darker.
Ocymyrmex monardi Santschi, 1930
T L 7.4-8.1 mm. Originally 0. monardi was described as a new variety of 0. weitzeckeri Emery, 1892. Accord- ingly, the syntypes are from “Cakindo” and “Chimporor” at the Rio Cubango (Okavanko) in Angola. Further material was mentioned by Bol- ton (1981) from Swakopmund, Namibia. The present samples from Mile 46 Research Station, Toggekry and Otjiamongombe confirm a prefer- ence for vegetation rich ecosystems. Considering also the record from Swakopmund this species can also be classified as mesotopic.
The colour of head and mesosoma is reddish- brown, gaster blackish, petiole, postpediole and gaster are basally tinted reddish.
Ocymyrmex picardi Forel, 1901
T L 11.6-12.6 mm. According to Forel (1901) the locus typicus is situated in Mossameded, Angola, between the rivers Rio Cubango and Rio Cuito. Further re- cords exist from different vegetation types in Botswana, Namibia, South Africa and Zimbabwe (Bolton 1989).
In this study, this mesotopic species was only recorded in Mile 46 Research Station. This con- firms a high preference for vegetation rich eco- systems.
This large and conspicuous Ocyrnyrrnex-spe- cies is best suited for monitoring studies in the field. As the habitat of this species is known for Namibia, bionomic data can be collected easily.
Colour of head dark reddish-brown, mesoso- ma and gaster black.
Ocymyrmex barbigepgroup
Ocymyrmex dekerus Bolton & Marsh, 1989
TL: 5.5-6.0 mm. The locus typicus of 0. dekerus is Grunau in southern Namibia. Further records exist from
the northern Cape Province (Bolton & Marsh 1989).
In this study this species was recorded in the more arid regions of southern Namibia and the Richtersveld in the Northern Cape Province. The new records confirm the large thermophily of 0. dekerus and provide important information to western distribution boundaries.
Colour of the head and gaster yellowish to light-brown, the gaster sometimes lighter, the mesosoma reddish.
Tetramorium Mayr, 1855
Worldwide, excluding the Neotropic region, 328 valid species are known. The centre of distribu- tion with 176 endemic species is in the Afrotro- pica1 region (Bolton 1980). So far 16 species were recorded from Namibia.
Tetramorium-species build their nests in rotten wood, under leaf litter, under roots, in tree holes as well as subterraneous. According to Prins (1978) there are two exceptions, 7: jordani Santschi ( T . aspinatum) and T glabratum Stitz (7: rutilum Prins), which completely lack or have reduced propodeal spines. Both species are re- stricted to the sandy coastal soils in Namaqua- land where they live exclusively subterraneous.
Tetramorium-species are generalistic feeders. Some species hunt as predators, while others har- vest grass seeds. Further, a few species are known, which live in ant-homopteran associations and col- lect honeydew of Aphidina or Coccina.
In this study, only those species that belong to the Tetramorium solidum-group and the T seri- ceiventre-group were recorded.
Tetramorium solidum-group
Teiramorium glabratum Stitz, 1923
TL: 5.0-5.7 mm. According to Prim (1978) this species occurs only in the desert regions of the western Nama- qualand along the coastline (Prins 1978). He re- fered to T rutilum Prins, 1973 which was synony- mized with T. glabrutrum by Bolton (1980). Locus typicus of T. solidum st. glabratum Stitz is Karibib, Deutsch-Sudwestafrika [Namibia]. Kari- bib is situated in the Western Highlands where the vegetation is formed by grassland and scat- tered trees (Mendelsohn et al. 2002), showing that T glabratum is not as strictly stenotop as formerly assumed.
Mitt. Mus. Nat.kd. Berl., 2001. Reihe 80 (2004) 2 269
Contrary to all expectations, T glabratum was not recorded from the central Namib Desert (Marsh 1986a, b), despite occurrance in Luderitz (Stitz 1923). The location presented here, Quag- gafontein in the Succulent Karoo is within the known distribution area of this species.
Colour reddish-brown, dorsum of mesosoma without long standing hairs, propodeal spines short, broad across the base and acute apically.
Tetramorium grandinode Santschi, 1913
TL: 5.0-5.3 mm; Fig. 4. This species has been found only in South Africa for a long time. Marsh (1986a, b) recorded it first from Namibia during his surveys east of the Namib Desert. Tetramorium grandinode is diur- nal and granivorous.
Colour blackish-brown, propodeal spines elon- gate and strong, postpediole in dorsal view enor- mously expanded laterally, dorsum of mesosoma without long erect hairs.
Tetramorium rufescens Stitz, 1923
TL: 4.0-5.1 mm. This species is only found in Namibia, and was recorded from the Fish River Canyon (Bolton 1980). In the Namib Desert on the gravel plains, nests are in small hollows. Tetramorium rufescens is diurnal and granivorous (Marsh 1986a).
Colour dull red, gaster darker than head and mesosoma, propodeal spines acute, strongly de- veloped, body without appressed glittering silv- ery hairs and long erect hairs.
Tetramorium setulifencm Emery, 1895
TL: 4.4-6.0 Tetramorium setuliferum is one of the most abun- dant ant species of southern Africa. It occurs in Botswana, Lesotho, Malawi, Mozambique, Nami- bia, Tanzania, Zambia and Zimbabwe. In Nami- bia it was only recorded in Gobabeb so far (Bolton 1980).
Colour dull red, propodeal spines elongate and strong, body with appressed glittering silvery hairs
Tetramorium sericeiventre Emery, 1877
TL: 3.3-4.4 mm. This species was recorded from nearly all coun- tries of the Afrotropical region. Further it is dis-
tributed in the African mediterranean countries as well as in Saudi Arabia and the Madagascan subregion. Tetramorium sericeiventre is an eury- topic species, which was confirmed for savannas in Tanzania as well as for open woodland (Ro- bertson 1998). In Namibia, this diurnal species was found in gravel plains and on interdune val- leys in the dune field of the Namib Desert. Here this omnivore species feeds on arthropod frag- ments, grass stems, leaves and seeds (Marsh 1986a).
Colour dull red, gaster black, propodeum armed with a pair of spines, body without ap- pressed glittering silvery hairs but with very few long standing stout hairs.
Tetramorium bicarinata-group
Tetramorium erectum Emery, 1895
TL: 4.2-5.0 mm. Tetrarnorium erectum had only been found in South Africa. The specimens from Karios in southern Namibia were the first recorded for Na- mibia.
Colour blackish-brown, propodeal spines elon- gate and strong, and all dorsal surfaces of body with long standing hairs.
Meranoplus E Smith, 1853
Up to now, 47 valid species are described for this genus. They are distributed with eight species in the Afrotropical region, with two species in the Madagascan subregion, with seven species in the Oriental region and with 30 species in the Aus- tralian region. From Namibia so far only Mera- noplus inermis Emery has been collected at “Okahanja” (Bolton 1981b). Current knowledge suggest that species of this genus - different from other genera - avoid the arid ecosystems and winter rain areas of southern Africa. The only exception is the obviously eurytopic M. peringueyi Emery, whose distribution area ranges from the Cape of Good Hope to the Natal-Dra- kensberge (Bolton 1981b).
The Afrotropical Meranoplus-species build their nests subterraneous and are often situated between roots. The nest entrance is craterlike (Bolton 1981b). According to Brown (2000) they are seed harvesters and general foragers.
Within this study two species were collected which are first time records for Namibia.
210 Koch, F & K. Vohland, Ants along a southern African transect
Meranoplus spiniodis-group
Meranoplus glaber Arnold, 1926
TL: 2.3-3.1 mm; Fig. 5. This species formerly was only found in Zim- babwe and Botswana. The finding in Mile 46 Re- search Station was the first record of this species for Namibia.
Colour pale yellow.
Meranoplus magrettii-group
Meranoplus magrettii And&, 1884
TL: 2.8-4.3 m. This species belongs to the most widely distribut- ed Meranoplus-species of the Aftrotropical re- gion. Records exist from Botswana, Ghana, Ken- ya, South Africa, Sudan, Tanzania, Uganda and Zimbabwe. For Namibia it is the first confirma- tion.
According to Bolton (1981b) this species pre- fers savanna and open woodland habitats in east- ern and southern Africa. The locality in the thornbush savanna of Namiba confirms the habi- tat requirements. In contrast to M . magrettii, M glaber seems to prefer regions with higher an- nual precipitation and could only be collected in north-eastern Kalahari woodlands.
Colour of mesosoma dull red, head and gaster dark brown to blackish.
Ph eidole Westwood 1841
This genus encompasses worldwide 910 valid species, with main distribution in the tropics and subtropics. From the Afrotropical region 115 spe- cies were described and from the Madagascan subregion 19 species. For Namibia only two spe- cies were known, which were confirmed within this study.
The workers are dimorphic, with two discret castes. The major workers have large heads, and are known as “soldiers”. But this name is mis- leading, because the soldiers are only seed crushers. Nests are mostly subterraneous or sometimes in rotten wood.
Pheidole-species belong to various trophic types. Most species are seed harvesters (Brown 2000). Other species are omnivor, honeydew fee- ders (Marsh 1981b) or predators.
The taxonomy of this genus is very compli- cated as no modem revisions exist for Ethiopian
species. Therefore information about distribution areas remains fragmented.
Pheidole robusta Stitz, 1923
TL: 2.0-4.5 mm; Fig. 6. Originally this species was described as a new variety of Z? tenuinodis by Stitz (1923) with place of origin from Swakopmund, Deutsch-Sudwest- afrika [Namibia]. Pheidole robusta was not men- tioned by Marsh (1986a, b), because probably he was unable to determine it. In this case, identifi- cation was only possible because the animals col- lected at Karios could be compared with type material from the MNHU.
Colour reddish-brown, legs yellow.
Pheidole tenuinodis Mayr, 1901
TL: 2.5-4.0 ITLIII. This species is the most widely distributed Phei- dole species in southern Africa. Marsh (1986a) reported it from the central Namib Desert up to the coast line (Marsh 1986b). Stitz (1923) men- tioned the localities Otavi and Okahandja.
The present results confirm a mesotopic distri- bution, as I? tenuinodis was collected in the thornbush savanna as well as in the Succulent Karoo.
Colour of mesosoma brown, head and gaster dark brown to black, legs dirty yellow.
Messor Forel, 1890
According to Bolton (1995) this genus comprises 162 valid species. The species are distributed in the Holarctic, Aftrotropical and Oriental region, with the centre of distribution laying in the Pa- laearctic region. In the Afrotropical region, 12 species are found plus one species from the Ma- dagascan subregion. For Namibia up to date, 5 species are known (Bolton 1982), with Messor leubberti Forel being the most widely distributed species ranging from East Africa to southern Africa. In the present study this species was not recorded despite its expected occurrence. The lo- cus typicus was Okahandja which is close to the biodiversity observatories Otjiamongombe and Togge kry.
The prefered habitats of these species are grasslands and savannas of arid and semi-arid re- gions. The absence of most Messor-species from the extremely arid ecosystems is probably re-
Mitt. Mus. Nat.kd. Berl., Zool. Reihe 80 (2004) 2 271
lated to the lack of appropriate vegetation as these species are seed harvesters. The nocturnal M. denticornis have been observed to collect parts of insects under certain conditions (Marsh 1986a).
The nests of Messor-species are subterranean and often contain many large storage chambers.
Messor-species are strongly polymorphic with an enormous range of worker size. Also in this case total length is very variable.
The material collected within the BIOTA study confirmed the presence of two species of this genus.
Messor denticomis Forel, 1910
TL: 5.5-11.0 mm; Fig. 7. Messor denticornis is known from Botswana, Na- mibia and South Africa. The localities in Nami- bia were in the Nama Karoo and Succulent Ka- roo (locus typicus: Liideritzbucht; Bolton 1982). Further, this species was recorded from several localities within the central Namib Desert (Marsh 1986a). The new locality Quaggafontein in the Succulent Karoo of South Africa belongs to the known distribution area in south-western Africa and again confirms the preference of M. denticornis for more arid regions.
Colour of head and mesosoma dark brown, gaster blackish, legs light brown.
Messor capensis (Map, 1862)
TL: 6.0-11.0 mm According to Bolton (1982) this species is known from Botswana, Namibia and South Africa. Type material was collected at Maltahohe and Betha- nien, Namibia. These localities as well as the new record from Gellap-Ost lie within the dwarf shrub savanna of the Nama Karoo.
A study (Vorster et al. 1992) indicates that a light grazing regime favors M. capensis.
Colour of the body black, legs black.
Subfamily Ponerinae
tropical region only the two species, 0. asmiensis Emery and 0. troglodytes Santschi, are known. One species, 0. coquereli Roger is described from the Madagascan subregion.
The nests are mostly subterranean under dif- ferent objects, in rotten wood or in leaf litter. Surface activity is nocturnal and diurnal.
They are priorily predators on small inverte- brates, hunting individually and capturing prey with their elongate trap-like jaws (Shattuk 1999). Sometimes trophobiosis seems to occur (honey- dew feeding) with Cixiidae (Auchenorrhyncha) (Holldobler & Wilson 1995).
In this study, only 0. troglodytes could be re- corded.
Odon tomachus troglodytes Sant schi, 1914
TL: 7.5-12.5 mm; Fig. 8. Originally, this species was described as a variety of 0. huematodus (Linnaeus) from the locus typi- cus “Westafrica”. Therefore, all distribution data from Prins (1978) was wrong. For the first time Brown (1976) separated 0. troglodytes as valid species from 0. haematodus which is distributed in the Nearctic region. Odontomachus troglo- dytes is distributed in West and East Africa as well as in southern Africa. It is absent from the southern and western Cape Provinces (Prins 1978).
Head and mesosoma dark brown, gaster black; pediole ending dorsally in a elongate, sharp spine.
Pachycondyla F. Smith, 1858
With the exception of the Palaearctic region, world-wide 201 valid Pachycondyla-species are known, which are distributed in the tropical and subtropical zones. From these, 53 species belong to the Afrotropical region and seven species to the Madagascan subregion.
There are no modern taxonomic revisions, hence the exact determination of remains uncer- tain.
Odontomachus Latreille, 1805 Pachycondyla cf. caffraria (F. Smith, 1858)
About 55 species of this genus are known (Bol- ton 1995b) with distribution in the Nearctic, Neotropic, Oriental and Australian region. Cen- tre of distribution are the Nearctic and Oriental region with 24 species, respectively. In the Afro-
TL: 8.0- 9.0 mm; Fig. 9 Locus typicus of f! cafSraria is South Africa. Spe- cimens assigned to P. cf. cafsrariu were collected at Toggekry and Otjiamongombe.
272 Koch, F & K. Vohland, Ants along a southern African transect
Subfamily Dolichoderinae
Tapinoma Forster, 1850
The workers of this genus are very small (TL: 1,5-3.0 mm); the node of the petiole is absent.
Up to now, 60 valid species are described with- in the genus Tapinoma, ranging from the tempe- rate zone to the tropics. About 13 species belong to the Afrotropical region, and four species to the Madagascan subregion (Bolton 1995b). There are no modern revisions, which makes species determination uncertain.
Tapinoma-species build their nests in a wide range of microhabitats, epiterranean and subter- ranean, between rocks, wood or other material, rotten wood and plant stems. Individual nests may either be small with very few workers or large with many thousands of workers (Shattuk 1999).
Tapinoma-species are generalized foragers, but they have a preference for honeydew collected from Aphidina and Coccina. The above-ground activity is nocturnal and diurnal (Shattuck 1999).
Within of the present study only one species with many individuals was collected. The abun- dance decreased from north to south. The spe- cies is missing in Gellap-Ost, which is remark- able, as in the directly bordering Nabaos plot this species was trapped in high abundance. Gel- lap-Ost had a denser vegetation cover than the degraded Nabaos plot, and higher arthropod ac- tivity (Vohland et al. 2004).
Tapinoma spec.
TL: 1.5-1.8; Fig. 10. Probably this animals belong to a new species which is closely related to Tapinoma demissum Bolton, 1995 (it is a new name for 7: gracilis Forel, 1913; praeocc. in Formica by Fabricius, 1804).
South of the Oranjerivier no Tapinoma-species could be recorded.
Conclusion
Ants can be useful monitoring organisms for environmental changes e.g. along the BIOTA transect. However, additional knowledge is required on biosystematics and ecological re- quirements. The observatories are a suitable tool, but they have to be analysed at a more detailled microhabitat scale, based upon botanical and pedological information from the pilot phase results.
Acknowledgement
The study was supported by the BIOTA program of the Ger- man Federal Ministry of Education and Research, project ID: OlLC 0024. Subproject leader was Prof. Ulrich Zeller from the Museum fur Naturkunde, Berlin. His engagement in the project is gratefully acknowledged.
Thanks to Renate Austermuhle for her help in determin- ing plants, Andreas Petersen for providing information about soil characteristics of the BIOTA plots, and Bianca Horsch in providing remote sensing information.
Also many thanks to the owner of the plots to allow and support these studies, namely the Ministry of Agriculture, Water and Rural Development; Pommersche Farmgesell- schaft with Renate & Horst Austermuhle as well as Vanessa Kaspar & Sebastian von Koenen; Elfi & Johan Kotze; Man- fred Goldbeck from Gondwana Canyon Park also for free logis; the Richtersveld National Park; Kowie Geldenhuis; Meent Commitee of Soebatsfontein; and the local communi- ty from Nuwe Fontein. The Ministry of Environment and Tourism, Namibia, the Northern Cape Nature Conservation Service, and the Richtersveld National Park, South Africa are thanked for providing permits and interest in the results.
We thank Dr. A. McArthur and Dr. B. Seifert for critical reviewsof this manuscript.
References
Agosti, D., Majer, J. D., Alonso, L. E. & Schultz, T. R. 2000. Ants. Standard methods for measuring and monitoring biodiversity. - Washington and London (Smithsonian In- stitution Press), xix + 1-280.
Alonso, L. E. 2000. Ants as indicators of diversity. In D. Agosti, J. D. Majer, L. E. Alonso and T. R. Schultz (eds): Ants. Standard methods for measuring and monitoring biodiversity, pp. 80-88. Washington and London (Smith- sonian Institution Press)
Arnold, G. 1924. Monography of the Formicidae of South Africa (Camponotinae). Part VI. - Annals of the South African Museum 14: 675-766.
Baroni Urbani, C. 1972. Studi sui Camponotus (Hymenopte- ra, Formicidae). - Verhandlungen der Naturforschenden Gesellschaft in Base1 82: 122-135.
Bolton, B. 1980. The ant tribe Tetramoriini (Hymenoptera: For- micidae). The genus Tetrumorium Mayr in the Ethiopian zoogeographical region. - Bulletin of the British Museum (Natural History), Entomology Series 40 (3): 193-384.
- 1981a. A revision of six minor genera of Myrmicinae (Hymenoptera: Formicidae) in the Ethiopian zoogeogra- phical region. - Bulletin of the British Museum (Natural History), Entomology Series 43 (4): 245-307.
- 1981b. A revision of the ant genera Merunoplus F. Smith, Dicroaspis Emery and Calyptomyrmex Emery (Hyme- noptera: Formicidae) in the Etiopian zoogeographical re- gion. - Bulletin of the British Museum (Natural History), Entomology Series 42 (2): 43-81.
- 1982. Afrotropical species of the myrmicine ant genera Cardiocondyla, Leptothorax, Melissotarsus, Messor and Cataulacus. - Bulletin of the British Museum (Natural History), Entomology Series 45 (4): 307-370.
- 1995a. A new general catalogue of the ants of the world. - Cambridge, London (Harward University Press). 1-504.
- 1995b. A taxonomic and zoogeographical census of the extant ant taxa (Hymenoptera: Formicidae). - Journal of Natural History 29: 1037-1056.
Bolton, B. & Marsh, A. C. 1989. The Afrotropical thermophi- lic ant genus Ocymyrmex (Hymenoptera: Formicidae). - Journal of Natural History 23: 1267-1308.
Brown Jr., W. L. 1976. Contributions toward a reclassifica- tion of the Formicidae. Part 6. Ponerinae, tribe Ponerini,
Mitt. Mus. Nat.kd. Berl., Zool. Reihe 80 (2004) 2 273
subtribe Odontomachiti. Section A. Introduction, subtri- bal characters, genus Odontomuchus. - Studia Entomolo- gica (N.S.) 19: 67-171. Diversity of ants. In Agosti, D., Majer, J. D., Alonso, L. E. & Schultz, T. R (eds): Ants. Standard methods for measuring and monitoring biodiversity. pp. 45-79. - Washington and London (Smithsonian Institution Press).
Elmes, G. W. 1991. Ant colonies and environmental distur- bance. In Meadows €! S. & Meadows A. (eds): The envi- ronmental impact of burrowing animals and animal bur- rows. Clarendon Press, Oxford. 15-32.
Forel, A. 1910. Formicidae. In Schultze, L. (ed.): Zoologische und anthropologische Ergebnisse einer Forschungsreise im westlichen und zentralen Sudafrika, ausgefiihrt in den Jahren 1903-1905, mit Unterstiitzung der Koniglichen Preussischen Akademie der Wissenschaften zu Berlin. Je- na (Fischer Verlag) 4: 1-30.
Hoffmann, A. & Zeller, U. 2003. Influence of different land use intensities on species diversity and abundance of small mammals in the Nama Karoo, Namibia. - Belgian Journal of Zoology (submitted).
Holldobler, B. & Wilson, E. 0. 1990. The ants. - Berlin, Hei- delberg (Springer Verlag). XI1 + 1-732.
Kaspari, M. & Majer, J. D. 2000. Using ants to monitor envi- ronmental change. In Agosti, D., Majer, J. D., Alonso, L. E. & Schultz, T. R (eds): Ants. Standard methods for measuring and monitoring biodiversity. pp. 89-98. - Wash- ington and London (Smithsonian Institution Press).
Koivula, M., Kotze, D. J., Hiisivuori, L. & Rita, H. 2003. Pitfall trap efficiency: do trap size, collecting fluid and vegetation structure matter? - Entomologica Fennica 14
Lindsey, P. A. & Skinner, J. D. 2001. Ant composition and activity patterns as determined by pitfall trapping and other methods in three habitats in the semi-arid Karoo. - Journal of Arid Environments 48: 551-568.
Majer, J. D. 1993. Comparision of the arboreal ant mosaic in Ghana, Brazil, Papua New Guinea and Australia - its structure and influence on arthropod diversity. In LaSalle, J. & Gauld I. D. (eds): Hymenoptera and biodiversity. pp. 115-141. - Wallingford, UK (CAB International).
1-14.
Marsh, A. C. 1986a. Checklist, biological notes and distribu- tion of ants in the central Namib Desert. - Madoqua 14
- 1986b. Ant species richness along a climatic gradient in the Namib Desert. - Journal of Arid Environments 11
Mayr, G. L. 1862. Myrmecologische Studien. - Verhandlun- gen der Kaiserlich-Koniglichen Zoologisch-Botanischen Gesellschaft in Wien 12: 649-776.
Mendelsohn, J., Jarvis, A,, Roberts, C. & Robertson, T. 2002. Atlas of Namibia. A portrait of the land and its people. - Cape Town (David Philip Publishers; New Africa Books
Prins, A. J. 1978. 27. Hymenoptera. In Werger, M. J. A. (ed.): Biogeography and ecology of southern Africa. pp. 823-875. - The Hague 2 (Junk Publishers).
Robertson, H. & Zachariades. C. 1997. Revision of the Cum- ponotuus fulvopilosus (De Geer) species-group (Hyme- noptera: Formicidae). - African Entomology 5: 1-18.
Shattuk, S. 0. 1999. Australian ants. Their biology and identi- fication. In Monographs on invertebrate taxonomy 3 XI + 226. - Collingwood (CSIRO Publishing).
Stitz, H. 1923. Beitrage zur Kenntnis der Land- und SiiBwas- serfauna Deutsch-Siidwestafrikas. In Michaelsen, W. (ed.): Ergebnisse der Hamburger deutsch-sudwestafrikanischen Studienreise 1911. Hymenoptera 7. Formicidae: 143-167. Hamburg.
Tshiguvoho, T. E., Dean, W. R. J. & Robertson, H. G. 1999. Conservation value of road verges in the semi-arid Ka- roo, South Africa: ants (Hymenoptera: Formicidae) as bio-indicators. - Biodiversity and Conservation 8: 1683- 1695.
Vohland, K., Uhlig, M., Marais, E., Hoffmann, A. & Zeller, U. 2004. Impact of different grazing systems on arthro- pods in Namibia, with special reference to beetles (Co- leoptera). - Journal of Applied Entomology (submitted).
Vorster, H., Hewitt, P. H. & Westhuizen, M. C. v. d. 1992. Nest density of the granivorous ant Messor capensis (Mayr) (Hymenoptera: Formicidae) in semi-arid grass- lands of South Africa. - Journal of African Zoology 106
