INTRODUCTION CHAPTER 1

INTRODUCTION CHAPTER 1

CHAPTER 1

INTRODUCTION

1.1 Background

On 5 June 1992, at the United Nations Conference on Environment and Development (the

Rio 'Earth Summit'), the Convention on Biological Diversity [CBD] was opened for signature (Davis et al. 1994). The CBD was inspired by the world community's growing commitment to sustainable development. It represents a dramatic step forward in the conservation of biological diversity, the sustainable use of its components, and the fair and equitable sharing of benefits arising from the use of genetic resources.

What is biological diversity? Biological diversity, or biodiversity, is the sum of all species of animals, plants, fungi and micro-organisms, as well as the gene pool, evolutionary history and potential value of each species (Low & Rebelo 1996; Garcia et al. 2001).

Biodiversity is an integral part of ecosystems, ecological processes and landscapes, which is much appreciated (e.g. Kruger National Park) and needed (e.g. Maputaland) by a country's people.

Plant species are fundamental independent building blocks of biodiversity to which people can easily relate in terms of various values. It forms a most obvious biological unit that is of substantial use to people and thus forms the unit that is the most significant part of biological research. The importance of plant diversity may be summarised as follows: humans and most other animals are almost totally dependent on plants, directly or indirectly, as a source of energy through their ability to ,sequester carbon from the atmosphere and to capture the sun's energy via photosynthesis (Salisbury

&

Ross 1992). Humans are currently using tens of thousands of species of higher plants, and several hundred lower plants, for a wide range of purposes. To conserve and utilise this plant life sustainably, it is important to know what is being used and what should be conserved.

I

In recent years there has been much interest in documenting biodiversity (Stork 1993;

Blackmore 1996; Cresswell & Bridgewater 2000). A lot of focus has been placed on the development of strategies to monitor biodiversity in developing countries by training and making use of parataxonomists (Huntley et al. 1998; Basset et al. 2000; Danielsen et al.

2000). Such strategies became necessary where the achievements of initiatives to strengthen biodiversity conservation in developing countries became difficult to assess because most of these countries have no system for monitoring biodiversity. Furthermore the need for georeferenced specimen-label databases (botanical collections) has become increasingly important as tools for conservation actions (Soberon et al. 2000; Ter Steege et al. 2000;

Siebert & Willis 2000), which in turn, are an output of the strategies to monitor biodiversity.

Before biodiversity can be monitored, an inventory listing the natural resources, namely species, ecosystems and environmental factors, of a region is needed. Such an inventory will aid future monitoring strategies and management plans of large natural areas. Knowledge of a region's plant diversity forms the basis for understanding and managing its plant resources and environment. Concern for the survival of plant species has driven conservation policies and actions for many centuries. One of the main problems facing plant conservation in southern Africa, is the lack of sound information on which to base conservation strategies.

Traditional botanical conservation systems have evolved around species that are of practical use (e.g. food, fodder, timber, medicine, etc.). However, governmental conservation policies have focussed on species that are charismatic and/or of scientific interest (e.g.

Welwitschia mirabilis, succulents, forest trees, etc.). The conservation of rare species that are threatened or endangered has also become a major concern of late shared by governments, conservation organisations and individuals. Although these approaches differ, it is the concern for the survival of plant species that has been the driving force in conservation.

Conservation issues are even more critical in areas that are prone to rapid development.

This is often the case in mining areas where the economic value of the underground layers are more important than the above ground value of the biodiversity. South-central Africa, that is South Africa, Zimbabwe, Botswana, Zaire and Zambia, are some of the richest areas

2

for heavy metals (especially copper, cobalt, chromium, nickel, platinum and uranium) in the world (Brooks

&

Malaisse 1985; Coetzee 1985; Nriagu

&

Nieboer 1988). Poor economic growth, large populations and poverty have driven southern African governments to opt for the more popular and sought-after underground reserves of heavy metals as a much needed income. This has placed the associated biodiversity of these areas under threat.

1.2 Sekhukhuneland Centre of Plant Endemism

Inadequate definitions and inconsistent use of phytogeographic terms have led to confusion in southern African phytogeography (Van Rooy 2000). Due to this problem, and for the purpose of this study, a centre of endemism is demarcated as an area with a high concentration of taxa with limited geographic distributions (endemics). A centre of endemism is therefore defined as a concentration of taxa (usually at species level) in the geographic distribution area of an endemic element, which in this case is the Flora of

Sekhukhuneland.

It is important to note that, in the broad sense of the term, Sekhukhuneland is a hotspot. Hot-spots are geographic areas where centres of diversity and centres of endemism overlap and are threatened by habitat modification and transformation (Myers 1990). On a larger scale Sekhukhuneland is part of the hot-spot associated with the northeastern

Drakensberg Escarpment (Matthews 1991, Siebert 1998), which broadly relates to

Croizat's "Barberton Node" (Croizat 1965), and includes, among others, the Barberton and

Wolkberg Centres of Plant Endemism.

The Sekhukhuneland Centre of Plant Endemism [SCPE] is a phytogeographic (floristic) region (group of quarter degree grids of similar floristic composition) within a larger region.

Due to the occurrence of a phytogeographic (floristic) element (group of taxaendemics-of similar geographic distribution) it was classified as a centre of endemism (Van

Wyk & Van Wyk 1997; Siebert 1998).

3

The criteria which were adopted to select the study area as a Centre of Plant Endemism was based on the characteristics described by Davis

et al. (1994) for Centres of Plant

Diversity:

• The area is evidently species-rich, even though the number of species present may not be accurately known (Siebert 2000);

• The area is known to contain a large number of species endemic to it (Siebert

1998);

• The site contains an important genepool of plants of value to humans or that are potentially useful (Kritzinger 1992; Crookes

et al.

2000);

• The site contains a diverse range of habitat types (Siebert et al. 2002a);

• The site contains a significant proportion of species adapted to special edaphic conditions (Siebert

et al.

2001);

• The site is threatened or under imminent threat oflarge-scale devastation (Siebert

et al.

2002b).

In View of the current focus on global biodiversity, it is not surprising that the identification of centres of plant diversity and endemism has become a matter of great urgency and importance (Myers 1988; Wilson 1992). The international Convention on

Biological Diversity has focused renewed attention on the rapid global loss and degradation of natural ecosystems (Convention on Biological Diversity 1994). Recently the World

Conservation Union (!uCN) and World Wide Fund for Nature (WWF) recognised approximately 235 Centres of Plant Diversity worldwide which are of primary importance for the conservation of plant diversity (Davis et at. 1994). These centres are endemic-rich botanical sites of global conservation significance. No fewer than 14 of the 84 centres in

Africa are located in southern Africa. However, many other centres of local importance have not yet been explored or investigated in great detail. The Sekhukhuneland Centre of

Plant Endemism (Van Wyk & Van Wyk 1997; Siebert 1998) is such an area. It is a poorly studied, serpentine-related, floristic region located in the Northern Province and

Mpumalanga on the ultramafic rocks of the Rustenburg Layered Suite of the Bushveld

(Igneous) Complex (Siebert 1998). The SCPE lies directly adjacent and to the west of the

4

Wolkherg Centre of Endemism (Matthews et al. 1993), a local focus of endemism within the Afromontane Region. The Afromontane Region is one of the 84 sites (Site

Af67)

in

Africa that has been identified as a Centre of Plant Diversity and Endemism (Davis

et al.

1994 )-the southern parts which are known as the Drakensberg Regional Mountain

System.

1.3 Rationale and motivation

Southern Africa has the richest temperate flora in the world (Arnold

&

De Wet 1993). For a predominantly warm-temperate, semi-arid region, southern Africa is exceptionally rich in vascular plants. Southern Africa, defined as the region south of the Cunene-Zambesi Rivers, comprises over 30 000 species of flowering plants and ferns, including the whole of one of the world's six floristic kingdoms (the Cape Floristic Region or Fynbos Biome). The region also includes (Huntley

et al. 1998):

• Seventeen ofthe centres of plant diversity identified by the IUCN!WWF;

• Arid and semi-arid ecosystems (with half of the world's known succulents);

• Coastal, marine and freshwater ecosystems (RAMSAR and World Heritage

Sites);

• Forest ecosystems (most under some threat);

• Mountain ecosystems (e.g. Richtersveld, Drakensberg and Maluti Mountains).

South Africa plays host to an enormous diversity of plants which comprises almost 10% of the world's known flowering plants (Davis et al. 1994). According to Arnold & De Wet

(1993), the approximately 24 000 infrageneric taxa of South Africa is the highest number of native vascular plant species recorded for any country in Africa. In the world only Brazil,

Colombia, China and Borneo have more native vascular plant species than South Africa.

According to Davis et al. (1994), South Africa carries the most endemics (> 16 000 species) in Africa, resulting in a species endemism of approximately 70%. This percentage compares well with endemism rates on islands like Madagascar. Only a few regions have a higher percentage endemism than South Africa.

5

The above-mentioned statistics concerning southern Africa, and in particular South

Africa's rich and unique plant diversity, justifies the attention it receives in numerous scientific works concerning floristics (Cowling & Hilton-Taylor 1994; Davis et al.

1994;

Myers et al. 2000; Van Wyk & Smith 2001). However, despite this great botanical diversity, the flora ofthe region has not been inventoried, monitored and evaluated to its full potential. In addition many plant species are threatened with extinction, mainly through agricultural, mining, industry and urban activities. The 2001 IUCN Red Data List of

Threatened Plants lists more than 4 300 plant species as threatened in southern Africa

(Golding 2001a).

Sekhukhuneland is a region within southern Africa which boasts a rich temperate flora on a relatively small land area. For instance, in South Africa the province ofKwaZulu-Natal has an estimated 5 000 indigenous species in an area of approximately 91 000 km

2

The

Sekhukhuneland Centre of Plant Endemism is about 20 times smaller (4 000 km

2

) and has approximately 2 000 indigenous species. This figure is extraordinary if compared with islands in the world, namely New Zealand has 2 000 species on 268 000 km

2 and Hawaii has

2000 indigenous species on 16600 km

2

However, when the percentage endemism is compared it is obvious that speciation in Sekhukhuneland has been much restricted as open niches were filled by plant species from surrounding areas. The percentage endemism for

New Zealand is 80%, Hawaii 95% and Sekhukhuneland about 4%. Considering that a restricted area is compared with complete biogeographical zones, a more significant figure can be obtained for the SCPE if all endemic species of the South Africa, which occur in the study area, are considered.

In this case it will mean that 28% ofthe species in the SCPE are endemic to South Africa.

According to the Pretoria National Herbarium Computerised Information System

[PRECIS] (data obtained on

11103/1997),

approximately 3 000 of the infrageneric taxa recorded for southern Africa occur along the eastern Bushveld Complex (Siebert 1998).

Furthermore, it is estimated that 520 of these 3 000 species are endemic to the northern provinces of South Africa. Of these 520 endemics, it was determined that approximately 50 specificlinfraspecific taxa are specifically endemic to the ultramafic substrates of

Sekhukhuneland (Siebert 1998).

6

An overview of this region's stormy history explains why, to date, not much was known about the area's flora and vegetation. The Sekhukhuneland of today cannot stand apart from a changing Africa. In the twentieth century, Sekhukhuneland was included into the

Lebowa homeland during the apartheid era in South Africa. During this time very little research took place regarding the sustainable development and conservation of the environment. The political and social instability in the area resulted in all research being focused on social aspects and hence, the flora, which form such an integral part of the peoples lives and heritage, has been neglected. Presently, heavy demands on natural resources imposed by the rapidly growing mining industry and a burgeoning human population, calls for the wise future use and management of the SCPE's plant diversity.

People are depleting the available natural plant resources as they struggle to deal with their poverty and already large parts of Sekhukhuneland, which were known to be tree or thicket covered, are now laid bare (Kritzinger 1992).

One of the main problems facing plant conservation in Sekhukhuneland is the lack of sound information on which to base conservation strategies. A baseline inventory of ecological data became essential to supply authorities with the necessary information required to designate areas for the most appropriate forms of land-uses, and to formulate management plans for the protection and sustainable use of the region's native plant resources. Hence, the primary motivation for this study stemmed from the urgent need to highlight those areas of prime botanical importance that is prone to rapid loss and degradation of natural ecosystems due to unplanned and uncontrolled development. An adequate database of natural features is essential for effective land-use management and implementation (Kent & Ballard 1988; Bedward et al. 1992; Rhoads & Thompson 1992).

Until the beginning of 1998 no in depth research concerning the floristic uniqueness of the area has formally been documented. Sekhukhuneland should henceforth be recognised for its endemic flora and heavy metal soils. Broad hypotheses were formulated to test for uniqueness in the plant diversity of the Sekhukhuneland Centre of Plant Endemism:

Hypothesis 1.

The Sekhukhuneland Centre of Plant Endemism is characterised by plant communities specific to its heterogeneous environmental factors.

7

Hypothesis 2.

The Sekhukhuneland Centre of Plant Endemism exhibits a specific relationship between certain plant species and heavy metal soils.

Hypothesis

3.

The Sekhukhuneland Centre of Plant Endemism has a rich plant diversity, including endemic, near-endemic and threatened plant taxa.

The hypotheses were used to formulate a wider objective that fits into the national vision for South Africa, an

immediate objective that is in accordance with the missions of both national and provincial nature conservation agencies, and specific

objectives of the

thesis to test the hypotheses and to contribute to vegetation science:

Wider Objective

To contribute towards sustainable human development in the Sekhukhuneland region through the effective conservation and utilisation of plant diversity resources and their associated habitats.

Immediate Objective

To develop a detailed knowledge base of the plant species diversity and habitat within the plant communities of the Sekhukhuneland Centre of Plant Endemism, adequate to monitor and evaluate future activities for conservation, sustainable use and rehabilitation of botanical diversity, in the face of specific developmental challenges in the region and in response to the Convention on Biological Diversity.

Objectives of the thesis

The overall objectives of this vegetation study of the SCPE, as discussed in this thesis, is to document and:

• Describe the plant communities and associated habitats (Chapter 5-9)

• Investigate the plant-soil relationships on heavy metal substrates (Chapter 10)

• Assess the plant diversity, plant endemism and threatened plant taxa (Chapter 11)

8

1.4 Principle aims

• To classifY the vegetation ofthe study area and formally describe the plant communities;

• To identifY endemic, rare and threatened plant species, and to incorporate such taxa in sections dealing with conservation and vegetation;

• To comment on the distribution of endemic/near-endemic taxa,

In particular plant communities and environmental conditions;

• To investigate the most important environmental factors that might have an influence on vegetation and floristic diversity ofthe study area;

• To investigate the habitat diversity with specific emphasis on habitats containing taxa of conservation value;

• To comment on the floristic status of the area and to identifY biogeographical patterns and floristic affinities between plant communities;

• To identifY the boundaries of the Sekhukhuneland Centre of Plant Diversity and

Endemism;

• To produce various maps of the floristic and environmental factors of the study area;

• To investigate and assess the major threats to the region;

• To produce a detailed first inventory of plant species in the study area, supported by voucher specimens deposited in the H.G.W.J. Schweickerdt Herbarium;

• To emphasize the potential impact of mining and residential development on the floristic diversity of the study area, and to make recommendations concerning sensitive plant species and floristic areas that should be conserved;

• To propose management guidelines and strategies for the wise utilisation of resources, and the conservation of floristic diversity in the study area.

1.5 Layout of thesis

The layout of this thesis can be divided into three main parts, namely phytosociology

(Chapters 4-9), plant-soil associations (Chapter 10) and floristic analyses (Chapter 11)

Each of these chapters was prepared from baseline data of scientific papers that have and will be submitted to scientific journals for possible publication. In addition, a general

9

introduction (Chapter 1), the study area in a broad sense (Chapter 2) and detailed methods

(Chapter 3) was presented, with additional information on these topics given under

Background

in Chapters 4 to 9, and under Introduction in Chapters 10 and 11. An abstract is also provided for Chapters 4 (covering Chapters 4 to 9), Chapter 10 and Chapter II. All the discussions of the scientific papers presented as Chapters 4 to 11 are brought together as Chapter 12. This major discussion focuses on relevant issues with regards to the main hypotheses and principle aims of this thesis. Appendices present a graphical account of the element concentrations in the rocks, soils and plants of the study area, as well as a checklist of the plant species recorded for the Centre of Endemism.

10

CHAPTER 2

STUDY AREA

2.1 History

Sekhukhuneland was named after the Pedi chief Sekhukhune I (Raper 1987), who succeeded the previous chief of the Pedi, Sekwati, who died in September 1861 (Monnig

1967; Smith 1967). The Pedi are the people who historically lived on, and around, the

Leol0 Mountains of Sekhukhuneland (Pollock

et at.

1963). According to Monnig (1967), the Pedi settled in Sekhukhuneland in approximately 1650. Very little is known of their history between 1650 and 1800, but the first definite date established in the history of the

Pedi, was the day ofthe solar eclipse, the day Thulare died in 1824 (Quin 1959).

Chief Thulare is recalled as the greatest and most loved of all the Pedi rulers (Monnig

1967). He managed to build a mighty nation and thus entered upon the most prosperous period ofPedi history. After Thulare's death, the powerful Pedi Empire came to a fall as a result of disruptions and arguments between his sons (Monnig 1967). At the same time the

Matebele raided Pedi territory and killed Thulare's sons, except for two (Monnig 1967).

One of the sons, Sekwati, gathered what he could of the Pedi and fled, leaving behind a country devastated by the Matebele, who had completely denuded the country of all stock and grain.

Fortunately, Sekwati returned after an absence of four years, restoring the Pedi kingdom (Monnig 1967). Chief Sekwati was a very diplomatic leader with great military skills (Smith 1967). He managed to resist attacks from both the Swazi and the Zulu, which made him a popular leader (Monnig 1967). According to Otto (1934), Sekwati established peace with the Europeans and both parties recognised the Steelpoort River as the boundary between the Pedi country, and the Republic of the Transvaal. This peace negotiation provided the Pedi protection against the

Swaz~ which resulted in the development of a very powerful Pedi tribe at the beginning of 1860 (Otto 1934).

11

After Sekwati's death, Sekhukhune stole the chieftainship from his brother Mampuru and thus inherited a powerful nation from his father (Otto 1934; Monnig 1967; Smith

1967). Chief Sekhukhune I strengthened his nation with rifles and accepted warriors from other nations (Otto 1934; Quin 1959). Soon Chief Sekhukhune's warriors started harassing smaller tribes and European farmers living near the borders of Sekhukhuneland (Quin 1959;

Monnig 1967). In these areas Europeans and tribesman battled for residual highlands which stand free from disease amidst the hot lowlands. Bush cover and rough terrain in the lowlands also restricted movement by horse and wagon and made such environments inaccessible (Pollock et at. 1963). Thus, the Europeans and smaller tribes sought their guns to protect their land from the Pedi.

The harassment of the European farmers and smaller tribes, as well as the persecution of black Christians, led to the first Sekhukhune war in 1872 (Otto 1934; Quin 1959). The

European farmers drove the Pedi back into their own country and tried to upset

Sekhukhune's rule, but failed. The second war was initiated by the British Empire after their annexation of the Transvaal (Quin 1959; Smith 1967). Chief Sekhukhune indicated that he resented British authority and commenced raiding the country. British and Swazi forces defeated him in 1879 (Otto 1934; Smith 1967).

The Transvaal was retroceded to the European farmers in 1881 (Quin 1959). The Pedi, like the other tribal groups in the Transvaal, received certain rights to land during the

Pretoria Convention (Pollock et al. 1963). It was hoped that some security from encroachment would bring the wars of the defiant Sekhukhune of the Leolo Mountains to an end. However, when the final boundary fixing came about, the Sekhukhune tribal land was not big enough to achieve anything more than 'reserve' status (Pollock

et al. 1963).

Thus, the area, though a local borderland big enough to merit the name of Sekhukhuneland, did not give rise to international or provincial boundaries as happened in the British

Protectorates, such as Swaziland and Lesotho. In later years, Sekhukhuneland became part of the former Lebowa, which was given to the Pedi as a homeland during the apartheid era.

This remained until the establishment ofa full democracy in South Africa (1994).

12

2.2

Locality

The Sekhukhuneland floristic region is located to the west of the northeastern Drakensberg

Escarpment in the Republic of South Africa (Figure 1). For the purpose of this study, the name 'Sekhukhuneland', is not used to describe the demarcated area on the map of the

Magisterial Districts and Provinces (1994) of South Africa, but rather to indicate the much wider Sekhukhuneland Centre of Plant Endemism. The SCPE lies within, and across, the political borders of the Sekhukhuneland Magisterial District in the Republic of South Africa, because political boundaries, usual1y, have little to do with environmental and biotic tendencies. It stretches from the Northern Province into Mpumalanga, and include towns such as Roossenekal, Schoonoord, Steelpoort, Sekhukhune, Burgersfort and Mecklenburg

(Figure 1)

The study area is situated between 24°15' and 25°30' S latitude, 29°30' and 30°30' E longitude. It is located to the west of the northeastern Drakensberg Escarpment where it encompasses approximately 4 000 km

2

The layers of the Rustenburg Layered Suite of the eastern Bushveld Complex (Kent 1980; Visser et al. 1989) underlie the core area of the

SCPE; it is bordered by Highveld Escarpment to the south, Strydpoort Mountains to the north, the Steenkampsberg and Drakensberg to the east, and the Springbok Flats to the west. The study area incorporates 12 quarter degree grids [QDGs] as defined by Siebert

(1998)', with every quarter degree grid covering about 675 km

2

(Edwards & Leistner

1971). The study area (Figure 1) lies within the fol1owing QDGs:

Grid

Quarter Degree

2429

2529

2430

2530

BC;BD;DB;DD

BB;BD

AC;CA;CB;CC;CD

AA

13

The chosen QDGs exhibit certain characteristics which are absent from the adjacent ones. These characteristics are:

• The surface area is covered by a substantial percentage of ultramafic rock;

• The vegetation is predominantly Bushveld;

• The topography is mountainous;

• The area is situated in the rain shadow of the northeastern Drakensberg Escarpment;

• The winter months are relatively frost-free.

Ultramafic rocks also occur in adjacent QDG 2529 AD, 2529BC and 2529DB, but these are not part of the study, primarily due to the higher frost intensity in 2529BC and 2529DB (also not bushveld) and the locality of 2529 AD, which is wedged in between other geological substrates outside the rain shadow of the

Drakensberg Escarpment.

The included, 2430CD, is characterised by all, except one, of the necessary requirements. The absence of the most important requirement, ultramafic substrates, renders it inappropriate as a QDG of the SCPE.

However, this QDG is wedged between the Bushveld of the study area and the grassland-covered mountains of the Drakensberg and the Steenkampsberg. Thus, it is included so that its bushveld affiuity with ultramafic substrates may be investigated.

2.3 Physical environment

A literature survey was conducted to obtain existing information on the physical environment (topography, geology, soils and climate) of the SCPE (Erasmus 1985; Chief

Director of Surveys and Mapping 1988; Visser

et al.

1989; McVicar

et al.

1991; Weather

Bureau 1998). The information and maps (Figures 2 to 5) provided will hopefully assist environmental consultants, conservationists and researchers when conducting critical surveys of the vegetation of the region.

The physical environment comprises a complex of ecological factors that determine the vegetation of a region (Rattray 1963):

14

• Physiographic factors which are determined by the topographic features (aspect and slope), exposure, altitude, relief, ground water and geodynamic processes (erosion);

• Edaphic facors which include soil structure, texture, depth, mineral composition, moisture availability, pH and aeration;

• Climatic factors which include rainfall, light, temperature, humidity, wind and evaporation;

• Biotic factors which are living organisms such as humans, large herbivores, invasive plants and tire.

2.3.1 Topography

The study area is known for its concentric belts of rocky ridges and mountains, and its traversing broad, heavily eroded valleys (Figure 2). Physiographically the study area can be described as a mountainous area that consists of rugged hills with flat to undulating valleys.

The topography of the SCPE is very heterogeneous and complex, a product of tectonic forces and magma surges 2 000 million years ago (Coetzee 1985), upon which the climate and erosive agents have promoted geomorphologic change (Marlow 1976).

The study area lies to the west of the north-south orientated northeastern Drakensberg

Escarpment that curves slightly westwards in its northern parts. Thaba Sekhukhune to the west of the Steelpoort River, the Strydpoort Mountains to the north of the Olifants River, the Steenkampsberg to the south and the Drakensberg to the west, isolates the study area from the Lowveld. It brings about what can be called a 'Iowveld' enclave on the Highveld.

As a whole the study area falls in the rain shadow of the northeastern escarpment with a resultant relatively low rainfall for an area located in the eastern part of southern Africa.

The Leolo Mountains and the Steelpoort River Valley are the most prominent topographic features in the SCPE, with either one or both of these features represented in seven of the 12 QDGs. From the Steelpoort River Valley, which lies at about 700 m above sea level [asl] (one of the lowest points), the Leolo Mountains rise to 1932 m asl (the highest point) (Chief Director of Surveys and Mapping 1988). In the central area it has a

15

topographical height asl variation of approximately 1 000 m over a distance of approximately 5 kIn.

The Leolo Mountains run from north to south through the centre of the study area; almost connecting the Strydpoort Mountains with the Thaba Sekhukhune Escarpment.

Several small mountain ranges occur in the study area -of these the Thaba Sekhukhune

Escarpment is the most diagnostic topographic feature. Other prominent mountains of the region include Tauteshoogte (1 789 m), Hoofstadkop (1 747 m), Morone (1 520 m),

Morole (1 403 m) and Phepane (1 436 m).

There are numerous valleys, from the Steelpoort River and Ohfants River Valleys and the valleys of their tributaries, to the areas between the various mountain ranges such as the

Leolo and Drakensberg ranges. These broad, flat valleys can be up to 60 kIn long and several in width. In general, the study area falls within the drainage basin of the Olifants

River. The east-flowing Olifants River and its tributaries sculptured the topography of

Sekhukhuneland over millions of years. Its headwaters captured a major west-flowing river after it had broken through the Great Escarpment during two major upliftments of the subcontinent, which occurred in the Miocene and late Pliocene (Partridge

&

Maud 1987).

Several rivers drain the basin, and flow through valleys averaging 750 m as!. The largest rivers feeding into the Olifants River are the Steelpoort, Lepellane, Moopetsi, Motse, Dwars

and

KIip

2.3.2 Geology

The lithology of South Africa can be divided into basic and acidic rocks. The difference between the two lies in the mineral content of the rock. Ultrabasic (ultramafic) rocks contain, for example, MgO, FeO and CaO, and acidic rocks contain mineral oxides such as

Si()" K

2

0 and Na

2

0 (Krauskopf 1967). Basic rocks are usually referred to as basalt. The

Bushveld Igneous Complex can be described as basaltic, because it contains relatively high concentrations ofMg, Ca, Fe, AI, and Cr compared to other rocks. Ultramafic rocks are not the result of weathering and subsequent sedimentation, but are products of the earth's mantle. These rocks are relatively 'new' and have only recently been exposed to weathering.

They are therefore not a normal inclination and are termed 'anomalous'.

16

Granite and shale are acidic rocks. Granite is known for its low mineral content and shale is formed from sediments derived from weathered rocks. These rocks are therefore not rich in heavy metals and are known as 'normal' rocks. Sediments derived from weathered rocks, such as shale, cover about 70% of the world's surface lithology

(Krauskopf 1967). Ultramafic rocks rich in heavy metals cover only a small portion of the earth's surface.

In contrast to most parts of the world, ultramafic rocks are plentiful in southern Africa

(Kent 1980; Roberts and Proctor 1992). Most of the world's economically exploitable deposits of heavy metals are located in the ultramafic rocks of South Africa, more specifically in the Rustenburg Layered Suite [RLS] of the Bushveld Complex (Coetzee

1985; Schiirmann

et al. 1998). Concentric belts of pyroxenite, norite and anorthosite of the eastern RLS commences near Stoffberg in the Mpumalanga province on the lower slopes of the Highveld Escarpment, continues northwards, crosses the Steelpoort River into the

Northern Province and then stretches northwards as far as the foothills of the Strydpoort

Mountains, a total distance of approximately 170 km. It has an average width of 30 km and is wedged between the Transvaal Sequence to the east and the Lebowa Granite Suite to the west (Visser et al. 1989). Some of the largest quantities of chromium and platinum in the world are present in the Critical Zone of the suite (Schurmann

et al. 1998; Viljoen &

Schurmann 1998). Chromium has been mined extensively in the past (Brabers 1970), but platinum even more so at present, due to very high market prices (Cawthorn 1999).

Since surface outcrops of ultramafic rocks of the RLS largely defines the area of the

SCPE, the geology is important and discussed in some detail (Figure 3). The Bushveld

Complex was formed during the Precambrian together with the Phalaborwa, Kunene and

Losberg Complexes, the Vredefort Granophyre and the Uitloop Granites, which represents the greatest mineral deposit event that has ever occurred on earth (Coetzee 1985). Before the formation of the Bushveld Complex, sedimentary rocks of the Transvaal Sequence covered the interior of what are today the northern Provinces of South Africa (age: 2 000 to

2 300 million years). Approximately I 950 million years ago a series of magma surges resulted in the emplacement oflava into the interior of the Transvaal Sequence as a result of alternating stress and pressure conditions (Visser et al. 1989). When the lava crystallised it

17

gave nse to different layers (Schiirmann et al. 1998). The tremendous weight of the congealed lava on the surface of the Transvaal Sequence resulted in its collapse. Layers of the Bushveld Complex and Transvaal Sequence were broken and exposed to the surface where it was weathered to its present state over millions of years.

Sekhukhuneland is known for its concentric belts (layers) of norite, which gave rise, among others, to the Leolo Mountains. The concentric belts in the SCPE, as they are visible today, are the exposed broken layers in their weathered state. This characteristic igneous layering of the Complex, is the product of crystallisation differentiation during successive surges of magma (Visser et al. 1989). The Provisional Tectonic Map of the Bushveld

Complex (Hunter 1975) clearly distinguishes between three exposed layers of the RLS. The three main groups of saucers (primary layers) that were crystallised are the Upper Zone, the

Main Zone and the Lower Zone (Kent 1980).

The RLS forms the outer limit of the Bushveld Complex, because it was deposited as the first (bottom) layers during the magma outflow (Keyser 1998; Schiirmann et al.

1998).

Its characteristic igneous layering is the product of crystallisation differentiation during successive surges of magma (Visser et al. 1989). The formation of the layers was dependent on the density of the minerals concerned (Kent 1980). When the lava reached the surface, the heavier metals sunk to the bottom where they crystallised first. The first layer that crystallised was the Lower Zone and is characterised by norite, bronzitite, dunite and serpentinised harzburgite as secondary layers. These layers contain main mineral components made up of elements such as Mg, Ni and Cr. The second saucer-shaped layer that crystallised was the Main Zone.

It is characterised by four predominant secondary layers namely, norite, anorthosite, pyroxenite and gabbro. These layers are characterised by mineral components rich in Ca, AI, Ti and V. The Upper Zone is characterised by two main secondary layers, namely ferrogabbro and ferrodiorite, and to a lesser degree, magnetite.

The main elements within the mineral components of these layers are Fe, Na, V and Ti. The crystallisation of chromites occurred between the Lower and Upper Zones and is referred to as the Critical Zone. The Critical Zone's secondary layers are mostly pyroxenite, norite, anorthosite, dunite and harzburgite. The main component of these layers contains rich quantities of Cr, Pt and Fe. Each of the layers can be further divided into secondary layers.

18

The secondary layers are distinguished from surrounding layers according to their main mineral component. The main mineral component consists of certain characteristic elements, e.g. olivine contains Mg and Ni and plagioclase contains Ca and Fe.

2.3.3 Soils

The predominant inclination of soils on earth is one of low heavy metal concentrations.

However, high concentrations of different heavy metals occur in metalliferous or serpentiniferous soils derived from ultramafic rocks (Kent 1980). The most naturally occurring ultramafic soils in Africa are those produced by outcrops of metal-bearing ores of copper, aluminium, nickel and iron (Wild 1978).

The heavy metal soils of the RLS are derived from gangue minerals such as norite, anorthosite, pyroxenite, gabbro, feldspar and rarely, magnetite (Coetzee 1985). These ganh'Ues are basaltic rocks and are the intermediate form between serpentine and granite

(Wild 1978). Basalt shows the highest concentration of selected elements when it is compared with granite and shale, and even the earth's crust (Krauskopf 1967). Granite gives rise to 'normal' soils and serpentine gives rise to 'toxic' serpentiniferous soils. Basalt contains higher concentrations of heavy metals than granite (Krauskopf 1967) and less than serpentine, and produces intermediate metalliferous soils (Wild 1978). Relatively high concentrations of heavy metals in the soils of Sekhukhuneland are therefore a consequence of its ultramafic origin.

The soils of South Africa consist of a very complex mixture of various types, and there are few cases where a single uniform type occurs over any large area (Figure 4). Mother material from which soils developed in Sekhukhuneland, are characterised by great variations in types, locality and abundance of elements (Hunter 1975; Marlow 1976). The

RLS holds some of the highest concentrations of heavy metals, such as Cr, AI, V and Ti, in the world (Coetzee 1985; Schiirmann

e/ al.

1998). The abundance of the elements varies from one area to another (see Chapter 5) and therefore the type of heavy metal soil occurring in a specific region is a result of the specific exposed layer of the RLS.

19

Ultramafic soils are considerably different from 'normal' soils in that they are rich in chromium, cobalt, iron, nickel and deficient in the nutrients calcium, molybdenum, nitrogen, phosphorous and potassium (Brooks 1987). The soils of the study area are typical for ultramafic areas, for they conform to the element composition and certain areas have a high

Mg/Ca ratio> 1 (Johnston & Proctor 1981). However, certain areas of the SCPE have a low Mg/Ca ratio

<

1, a phenomenon is presumably caused by a reduction in the binding strength of Mg due to topsoil acidification and subsequent Mg leaching (Roberts &

Rodenkirchen 1995). This explanation holds for the RLS, as soils of these strata have a variable pH (6-8) (Loock et al. 1982).

Physiography and climate, together with the underlying rock, determine the nature of the soils that are formed (Rattray 1963). Soil types of the SCPE are characterised by clays.

Ultramafic soils of the SCPE are mainly red or black montmorillonitic clays (Werger &

Coetzee 1978). These soils are vertic to melanic A-horizons and are rich in smectite clay minerals and ions such as Ca, K, Na, and especially Mg (McVicar et al. 1991). The soils are generally dark-coloured and occur in both upland and bottomland positions (Land Type

Survey Staff 1987; 1988; 1989). Prominent soils of this type identified for the SCPE are

Arcadia, Bonheim, Mayo, Milkwood and Steendal forms. Soils with ortic A-horizons and one of the following B-horizons, namely yellow apedale, red apedale, red structured, pedocutanic, neocutanic or lithocutanic, are also common in the SCPE. These include the following forms, namely Clovelly, Hutton, Shortlands, Valsrivier, Swartland, Oakleaf,

Mispah and Glenrosa.

Groups of soil types develop under similar conditions and four basic soil groups are distinguished for the study area (Monnig 1967). Most of the area consists of ferruginous

lateritic

soils, with broad intrusions of turf in certain areas. Towards the extreme east and south there are smaller intrusions of mist-belt soils. Large areas of the SCPE are characterised by low altitude soils below the footslopes of mountains .

• Ferruginous lateritic soils include grey lateritic, brown/dark brown ferruginous and deep red sandy-loam soils. It originates from both sediments and igneous rocks, and this, together with climate and natural drainage, determined the chemical and physical

20

qualities of the soil. Where the internal drainage is poor, the soil is stony and shallow

("Lesikihledi") with a solid lateritic base.

• Turf soils ("Seloko") consist of black and red clay types, of which the black is usually found in dry areas with inadequate drainage. These soils have a heavy, coarse texture and tend to crack to considerable depths. The top-soils slake to form a fine, granulated layer. These soils tend to retain moisture.

• The mist-belt soils developed in high altitude areas of high temperature and rainfall, which accelerated the soil forming process. This results in its formation on practically all the geological formations of the study area. These soil layers vary from red to yellow and have a clayish texture with a structure that gives an adequate internal drainage.

These soils are also intensely cultivated.

• Low altitude soils have better internal drainage. Gravel and stone tend to disappear and a deep soil with a sandy-loam to clay texture ("Mehlabane") results. The soils of low altitudes consist of many types, most of which can be fairly shallow. In these areas highly eroded basic igneous rock formations such as norite predominate. The soils are deep red to brown loam or clay soils of an excellent structure.

According to the Soil Degradation Index ofHoffinan & Ashwell (2000a), the southern part of the SCPE that lies within the Mpumalanga Province is rated to have light soil degradation. The northern part of the study area that falls within the Northern Province has

severe

soil degradation as measured with the SDI (Hoffinan & Ashwell 2000b). This index is based on soil erosion and overgrazing.

2.3.4 Climate

The study area lies in the summer rainfall region and the average annual rainfall for the

SCPE is 578 mm (South African Weather Bureau 1998), but the rainfall pattern is strongly influenced by the local topography (Siebert 1998) and varies from as little as 400 mm in some of the valleys, to an estimated 700 mm on the Leolo Mountains and in the extreme south (Mapochs Gronde) (Siebert 1998). Perhaps the most outstanding climatic feature of

21

;1 59yIIQ)l b iii"

)(01(,1'7

the central and northern parts of SCPE is that it lies in the rainshadow of the northeastern

Drakensberg Escarpment.

Sekhukhuneland receives nearly half its rain (48%) between December and February

(summer), an average total of 283 mm for these three months (Erasmus 1985). The peak month is January with an average monthly rainfall of 100-120 mm for the mountain bushveld areas and 140--{)00 mm for the temperate grassland areas. Throughout the country

May to August are generally dry. In the marginal months of April and September there is a general average rainfall of 20 mm per month. Spring rains that contribute 28% of the total rainfall in a single year usually precede the summer rains, but can stay away.

The rainfall gradient extends from southeast to northwest (Siebert 1998). The western part of the study area receives less rain on average than the eastern parts. There is a gradual increase in rainfall from west to east, with a sharp increase in the east, on the border with the Drakensberg foothills. Fluctuations can be attributed to altitude. The northern parts of the study area are also drier than the south. The north-central part of the SCPE is the driest, with the average annual rainfall for the study area increasing towards the Steenkampsberg that form the border in the south and the Strydpoort Mountains that form the northern extremity.

The whole study area has a fairly drawn-out warm summer, with a short mild winter.

January is generally the warmest month and July the coolest. Extreme temperatures for the study area range from -4.SoC to 38°C. The daily average is approximately 18.SoC (Weather

Bureau 1998). Taking the mountain bushveld and grassland areas respectively, the average minimum temperatures in January are 18°C and 14°C, and the average maximum temperatures 32°C and 26°C. The average minimum temperatures for July are 6°C and 2°C, and the average maximum temperatures 24°C and 20°C respectively. Temperatures vary at different localities within the study area, also correlating strongly with physiographic regions, being higher in low-lying valleys and lower on high-lying plateaus (Buckle 1996).

However, minimum temperatures of below freezing point are rare, even in the high-lying areas.

22

The northern and western parts of the study area are on average warmer than the south and east. The northern and western parts have average daily temperatures of 28.3°C maximum and 7.2°C minimum. These temperatures compare well with those associated elsewhere with Mixed Bushveld (Van Rooyen & Bredenkamp 1996). Average daily temperatures of the southern and eastern regions are more temperate and below those expected for Mixed Bushveld.

Temperature data also exhibit a set climatic pattern like that described for rainfall.

Valleys have a subtropical climate with no frost in winter, whereas in the mountains the conditions become more temperate with frost in winter as altitude increases towards the

Steenkampsberg. On the whole the study area is frost-free.

Lower rainfall in the western and northern parts of the SCPE correlates with the warmer temperatures in these parts (Siebert 1998). Climatically the SCPE comprises an arid

(karroid), subtropical (Iowveld) enclave surrounded by mountains that are temperate (frost in winter) and much wetter (particularly towards the east and south (Van Wyk & Smith

2001). The SCPE can be divided into climatic regions (Figure 5). This includes the

Bushveld areas with a dry, warm desert climate and a summer rainfall which is similar to the

Central-Western United States of America and the southern Russian Steppes (Monnig

1967) and can be divided into two climatic regions namely the (1) northern, moderately dry

(350-450mmlannum) and warm (21-22°C daily average) arid bushveld and the (2) central region, intermediate, typically Mixed Bushveld rainfall (450-600mmlannum) and temperature averages (20°C). The third (3) climatic region is the southern region, which is moderately wet (600-850mmlannum) and cool (18-19°C daily average), and can be described as a temperate, cooler escarpment zone with a typical Highveld climate that comprises dry winters and good summer rainfall (M 6nnig 1967).

2.4 Vegetation and flora

The most important works on the vegetation of South Africa are those by Pole Evans

(1936), Adamson (1938), Acocks (1953), Werger (1978), White (1983), Rutherford &

Westfall (1986), Low & Rebelo (1996) and Cowling et al.

(1997).

23

Rutherford & Westfall (1986) and Low & Rebelo (1996) identified eight biomes in

South Africa in accordance with dominance or co-dominance of plant life forms. The SCPE falls predominantly within the Savanna Biome and to a lesser extent includes an ecotone with the Grassland and elements of the Forest Biomes. Savanna is a tropical plant assemblage where the herbaceous stratum is continuous and prominent, interrupted to a greater or lesser extent by fire tolerant trees and shrubs (Lamotte 1985).

The Savanna Biome covers the greater part of the Northern Province and the northern parts of North-West Province. The area comprises mostly undulating to flat plains, at an altitude of 700 to 1 100 m above sea level (Van Rooyen & Bredenkamp 1996). Savanna is characterised by a grassy ground layer and a distinct upper layer of woody plants. When the vegetation has an upper layer near the ground, the vegetation may be referred to as

Shrubveld, where it is dense it is referred to as Woodland, and the intermediate stages are locally known as Bushveld (Rutherford & Westfall 1986). Fire and grazing is known to determine the structure of the Savanna Biome.

The vegetation maps by Acocks (1953) and Low & Rebelo (1996) are used as references for this study. The vegetation map of Acocks (1953) is the older version of the two (approximately 40 years) and is based on Veld Types. The map of Low & Rebelo

(1996) is based on Vegetation Types.

Acocks (1953) classifies the study area as three Veld Types, which includes the Mixed

Bushveld (18), Sourish Mixed Bushveld (19) and North-Eastern Sandy Highveld (57).

According to this vegetation map, Sekhukhuneland is bordered by the North-Eastern

Mountain Sourveld (8), Springbok Flats Turf Thornveld (12), Bankenveld (61) and to a lesser degree, the Lowveld Sour Bushveld (9).

According to the vegetation map of Low

&

Rebelo (1996), the QDGs of the SCPE are classified as one vegetation type, namely the Mixed Bushveld (18) Vegetation Type. On this vegetation map, Moist Sandy Highveld Grassland (38), North-Eastern Mountain Grassland

(43), and to lesser degree by Clay Thorn Bushveld (14) and Afromontane Forest (2) borders the SCPE.

24

The larger part of the SCPE was classified as Mixed Bushveld by both vegetation maps and hence will be treated as such. The Mixed Bushveld covers an area of 642 600 km

2

, of which approximately 3 500 km

2

(0.5%) occurs in Sekhukhuneland. The Mixed Bushveld is one of25 Vegetation Types recently defined for the Savanna Biome (Low & Rebelo 1996).

The Mixed Bushveld represents a great variety of plant communities, with many variations and transitions. The vegetation varies from a dense, short bushveld to a relative open tree savanna. On shallow soils Combretum apiculatum dominates the vegetation type.

Other trees and shrubs include

Acacia caffra, Dichrostachys cinerea, Lannea discolor,

Sc/erocarya birrea and various Grewia species. Here the grazing is sweet, and the herbaceous layer is dominated by grasses such as

Digitaria eriantha, Schmidtia

pappophoroides, Anthephora pubescens, Stipagrostis uniplumis and various Aristida and

Eragrostis species. On deeper, and more sandy soils, Terminalia sericea becomes dominant, with Ochna pulchra, Grewia flava, Peltophorum africanum and Burkea africana often prominent woody species, while Eragrostis pal/ens and Perotis patens are characteristically present in the scanty grass sward.

According to a survey done by Kritzinger (1992), the vegetation of Maandagshoek in

Sekhukhuneland differs from typical Mixed Bushveld and varies from open shrubland to dense bushveld. On shallow soil, covered with chalky gravel,

Eragrostis lehmanniana

dominates the vegetation, with species such as Diospyros lycioides var. guerkei and

Heteropogon contortus proving prominent. On the clay soils the sweet veld include species such as Fingerhuthia africana, Dichrostachys cinerea, Combretum hereroensis and

Hippobromus pauciformis. Rocky soils are characterised by Eragrostis rigidior, Psiadia

punctuata, Dichrostachys cinerea and Sclerocarya birrea occur. On the rocky dry hills

Aristida transvaalensis becomes dominant, with species such as Catha transvaalensis,

Acacia caffra and Elephantorrhiza praetermissa appearing diagnostic. Other prominent species of the region include Croton gratissimus, Vitex obovata subsp. wilmsii,

Enteropogon macrostachys and Rhoicissus sekhukhuniensis. Certain heavily eroded areas

(vegetation anomalies, though not serpentine) are very sparsely vegetated with distinctive flora, including Rhus keetii, Euclea linearis, Polygllia sp. nov. and Pterothrix spinescens.

25

Cole (1986) classifies the bushveld of the SCPE as part of the

Savanna Parklands and

associated law Savanna Woodlands that are typical for South Africa. The whole study area is savanna, with mixtures of grassland, but with bushveld predominating. The greater part of the study area can be described as a sweet bushveld. It has a thick, rich covering of various types of palatable graminoids, notably

Heteropogon contortus, Setaria sphacelata,

Themeda triandra and Tristachya leucothrix.

The bushveld in Sekhukhuneland is mainly a dense vegetation type and is characterised by deciduous trees, particularly Combretum apiculatum and

C.

molle, and other trees such as Terminalia prunioides, Kirkia wilmsii, Euclea crispa and various Acacia species

(Monnig 1967). The grass cover is thick with many herbaceous herbs.

The southern part of the study area lies in Mpumalanga and is classified as having

insignificant veld degradation on the Veld Degradation Index (VDI) ofHoffinan & Ashwell

(2000a). However, the northern part of the SCPE that lies within the Northern Province is rated to have

severe veld degradation, with the Schoonoord District being one of the top twenty districts in South Africa that requires priority attention in terms ofland degradation

(Hoffinan & Ashwe1l2000b).

Approximately 15% of the land area in the Northern Province and Mpumalanga has been invaded by alien plant species (Hoffman & Ashwell 2000a; 2000b). Invasive plants are a serious problem in the provinces where the study area is located and has influenced the current floristic composition of many systems.

2.4.1 Floristic history

The flora of the SCPE is mainly of Zambezian extraction, with Mromontane elements, especially at higher altitudes. According to the floristic map of White (1983), which indicates the main phytochoria of Africa and Madagascar, Sekhukhuneland is located within the Sudano-Zambezian Region, or more precisely its Zambezian Domain (Zambezian

Regional Centre of Endemism), on the border between the former and the Mromontane

Archipelago-like Regional Centre of Endemism. The archipelago is spread over southern

26

Africa, mainly along the eastern escarpment, but also in the south, reaching the Indian

Ocean coast (Werger & Coetzee 1978).

The Zambezian Regional Centre of Endemism covers virtually the entire high plateau of southern Africa and comprises vast stretches of woodland, savanna and grassland vegetation with occasional dry forests and thickets, and patches of swampy vegetation

(Werger & Coetzee 1978) Over large parts of the enormous area covered by the

Zambezian Regional Centre of Endemism, the rich flora only gradually changes, possibly as a result of the lack of strong relief and other contrasting physiographic factors (Werger &

Coetzee 1978). The Zambezian phytochorion probably has the richest and most diversified flora in Africa (White 1983) and stretches across ten countries. It emphasises how important it is to gather data on diversity and endemism by means of floristic provinces and not by political subdivisions.

On this larger scale, the Mixed Bushveld Vegetation Type of Rebelo & Low (1996) would be classified as part of the

'South Zambezian undifferentiated woodland and scrub woodland'

in the Zambezian Domain of the Zambezian Regional Centre of Endemism

(White 1983). In structure and floristic composition it is intermediate between

'North

Zambezian undifferentiated woodland' and 'Tongaland-Pondoland semi-evergreen

bushland and thicket'. Half of the recorded species in the 'South Zambezian

undifferentiated woodland and scrub woodland' are widespread in the Zambezian Domain

(White 1983). The remainder are more or less confined to the southern fringes of the

Zambezian Domain, which are mostly situated in South Africa, and include the SCPE.

The southern fringes are recognised by certain taxa which characterise the southern element, for example (White 1983):

Acacia caffra, Aloe arborescens, Grewiaflava, Kirkia

wilmsii, Protea caffra, Ptaeroxylon obliquum, Rhus leptodictya, Schotia brachypetala and

Spirostachys africana (White 1983). Many of these are shrubs or small bushy trees. Some are deciduous and others are evergreen. All the above-mentioned diagnostic species are present in the SCPE.

27

Although the vegetation of the study area is mainly of Zambezian extraction,

Afromontane links are to be expected as the region abuts on the northeastern Drakensberg

Escarpment with a mainly Afromontane flora (Wolkberg Centre). The vegetation of the

Afromontane Archipelago-like Regional Centre of Endemism mainly consists of dense forests, but also contains grasslands and savannas (White 1978). The most extensive vegetation type existing today in the Afromontane is fire-maintained grassland (White

1978). The change from Afromontane to Zambezian is particularly noticeable when one descends from the high-lying, wetter, more temperate Steenkampsberg to the low-lying, much drier, subtropical Roossenekal-Steelpoort area. Thus, many taxa of the SCPE are shared between the Zambezian Regional Centre of Endemism and the Afromontane

Archipelago-like Regional Centre of Endemism (Siebert 1998).

White (1983) also mentions the Zambezian flora on heavy metal soils, but only recognises such phenomena outside the boundaries of South Africa. Toxic amounts of heavy metals in the soil break the uniformity of the prevailing woodlands in the Zambezian

Domain with areas with very sparse vegetation. The less heavily contaminated soils support an open bushland or wooded grassland. This is also characteristic in the SCPE, however the species composition differs.

28

MOUNTAINS

SOUlH AFRICA

Sekhukhune

Roossenekal

STEENKAMPSBERG

PUMALANGA

Dullstroorn

o

1 : : 1

201cm

==I::::=:::::J'

Figure 1 Location of the Sekhukhuneland Centre of Plant Endemism in the Northern

Province and Mpumalanga, South Africa (based on Van Wyk & Van Wyk (1997), Siebert

(1998) and Van Wyk & Smith (2001».

29

-

STEENKAleSBERG

N t

o

211111n

1 : : : : '

=::::::I:===l'

Figure 2 Topography of the Sekhukhuneland Centre of Plant Endemism (based on Van W yk

&

Sm ith (2001» .

30

_ _ _

Vsi

.

_.-"_

...

: ••••••••••• " •

--."--.-..

-----....

-'.

-----. Vdr

-·0 .. ,

.....

\

.....

\

Vsi

i

Vdj

.

"

.

.:

.

'

.

..... .

Vrs ....

,,_

..

'.

Sekhukhune ...

:Vmc·· ..

:. Mn

..

.

........

-."

"

:

". Vmb.:

-.'

". • ·'.Vm

-0 . ' .

....

\.~c;·\ .

. .

'..

. .

.

: Vdr:. Vve

Vdj' :

.

~'

..

-

--

.... -

.--

-',

... Vcr • Burgersfort

..

,

i""

.

Vsn

..

~vsti··

....... :· o

1 = 1

201m

==±===="

Vdj

:

'

.

Lydenburg.

Roosseneka/: •

Vrs

Transvaal Sequence

Vmb

=

Magaliesberg Formation: quartzite, shale & homfels

Vmc

=

Mackekaan Formation: quartzite, arkose & sandstone

Vsb

=

Steenkampsberg Formation: quartzne, shale

& sandstone

Vsi

=

Silverton Formation: quartzite, shale, hornfels, limestone & dolomite

Vve

=

Vermont Formation: quartzite, hornfels, limestone, sandstone & chert

Bushveld Complex

Mn = Nebo granite

Rustenburg Layered Suite

Vrs

=

Roossenekal Subsuite (Upper Zone): mainly ferrogabbro, magnetite & norite

Vdj

=

Dsjate Subsuite (Main Zone): mainly norile, gabbro

& anorthosite

Vdr

=

Dwars River Subsuite (Critical Zone): mainly pyroxenite, norite & anorthosite

Vcr

=

Croydon Subsuite (Lower Zone): mainly pyroxenite

& norite

Vsn = Shelter Norite: mainly pyroxenite

& norite

Vis

=

Melanorite, pyroxenite

& serpentinized harzburgite

Figure 3 Major geological substrates of the Sekhukhuneland Centre of Plant Endemism (based on

Marlow (1976), Kent (1980), Visser

e/ al.

(1989) and Keyser (1998)).

31

I o

20km

1::1 =::::1'==:::11

Lydenburg.

Ac

=

Red-yellow apedal, freely drained soils; red and yellow dystrophic and/or mesotrophic

Ae

=

Red-yellow apedal, freely drained soils; red, high base status, > 300 mm deep (no dunes)

Ah

=

Red-yellow apedal, freely drained soils; red and yellow, high base status, usually < 15% clay

Dc

=

Prismacutanic and/or pedocutanic horizons dominant, wtth one or more vertic, melanic and red structured diagnostic horizon

Ea

=

One or more vertic, melanic and red structured diagnostic hOrizon, undifferentiated

Fa

=

Glenrosa and/or Mispah forms (other soils may occur); lime rare/absent in the entire landscape

Fb

=

Glenrosa and/or Mispah forms (other soils may occur); lime rare or absent in upland soils, generally present in low-lying soils

Ib

=

Miscellaneous land classes; heterogeneous rocky areas with miscellaneous soils

Ie

=

Miscellaneous land classes; rock wtth little or no soil

Figure 4 Major soil patterns of the Sekhukhuneland Centre of Plant Endemism (based on McVicar et al.

(1991) and Land Type Survey Staff (1987, 1988, 1989».

32

500

450

21.8

.'

.

22.6

400

Sekhukhune.

• Burgersfort

N

1

o

20km

1 : 1

==1::::::=;::11

21.8

18.3

Lydenburg.

22.6

I

Rainfall isohyet (mmlyr)

.-

.-.-

Temperature isohyet

(average daily ·C/yr)

Figure 5 Major climatic patterns of the Sekhukhuneland Centre of Plant Endemism (based on

Erasmus (1985), Siebert (1998) and Weather Bureau (1998)).

33

CHAPTER 3

METHODS

3.1 Phytosociological assessment

Experimental research in ecology is difficult to carry out, especially at the more complex levels of communities, ecosystems and landscapes. Most studies at these levels, such as the work presented in Chapter 4, are descriptive. The sets of data being analysed are large; they are gathered during field surveys. The analytical techniques used are determined by the objectives of the project; the results are influenced by what is sampled and the way it was carried out (Jongman et

al. 1995).

Quantitative approaches or numerical techniques have been used extensively in plant ecology and phytogeography. Examples of quantitative approaches at the fine-scale vascular plant community or phytosociologicallevel in southern Africa include studies by Coetzee et

al. (1995), Richards et al. (1995), Brown et al. (1996), Van Wyk et al. (1996), Witkowski

& O'Connor (1996), Smit et al. (1997), Sullivan & Konstant (1997), Cilliers &

Bredenkamp (1998), Kirkwood & Midgley (1999), Matthews et al. (1999), Lechmere-

Oertel & Cowling (2000) and Van Wyk et

al. (2000).

The non-statistical Braun-Blanquet method, as described by Mueller-Dombois

&

Ellenberg (1974), Werger (1974) and Westhoff & Van der Maarel (1982), was used to classity the vegetation of the SCPE into homogeneous physiognomic-physiographic units.

In an analytical phase the environmental, floristic and structural data are collected in the field. The data are then classified in the synthetic phase, to deliver the delineation of plant communities on the basis of their floristic and structural differences.

The current study has moved away from the methodology and technique and focussed more on the ecological application. This deductive approach uses phytosociology as a tool rather than an end in itself.

34

3.1.1 Analytical phase

The analytical phase was conducted over two growing seasons in 1999 and 2000, from mid-

December to mid-April. Initial reconnaissance surveys were done on several occasions prior to December 1999 to become familiar with the patterns in the climate, geology, topography and vegetation of the area. Voucher specimens of conspicuous plants were collected throughout the study period and were identified by the candidate and staff of the HG.W.J.

Schweickerdt Herbarium [PRU), University of Pretoria and the National Herbarium [PRE),

Pretoria.

The study area was stratified into relatively homogeneous physiographic-physiognomic units by recognising and mapping possible uniform vegetation units from aerial photographs

(1:25000) and further assisted with geological maps (1:50000). This enabled a sound stratification of the study area for efficient sampling of the representative vegetation types.

Stratification was also based on terrain type and aspect. Sample plots were placed in such a way that the habitat was as uniform as possible within each vegetation stand. Homogeneity is difficult to test statistically, therefore it was assessed visually and care was taken not to place plots in ecotonal zones.

Subjective stratified sampling (partially random) was used to allocate sample plots to physiographic-physiognomic units. The number of sample plots per unit depended on the unit size. This ensured that no over or under sampling occurred in such a large area of

4 000 km

2

In the field the sample plots were placed randomly within each homogeneous physiographic-physiognomic unit. The number of plots for the study area depended entirely upon the scale of the survey and subsequently it was decided to use a minimum of five releves per homogeneous vegetation unit (Hin 2000).

Plot size was standardised at 400 m

2

(20 x 20 m) for both the savanna, grassland and forest areas to counter the bias of different scale (Jonsson & Moen 1998). This large plot size was chosen due to the large scale of the survey and the heterogeneity of the SCPE vegetation. A large plot size ensures that a more representative model of the SCPE is investigated during the short time given for the study. Plots were shaped as to conditions to enhance homogeneity. Where the pre-determined position of the sample plot did not meet

35

the requirements mentioned above, or fell on structures such as roads, the plots were moved to more suitable areas. It must be noted that phenological changes over the year influenced the species cover and consequently the data of the releves (Fischer 2000).

In the Braun-Blanquet method a complete species list of vascular plants is normally compiled for each stand to derive a comprehensive floristic description. This requirement cannot always be met in semi-arid areas with unpredictable rainfall, and because multiple visits to a sample plot is not possible due to the limited time available and the extend of the study area. Therefore an attempt was made to compile as complete a species list as was possible in the time available. Plant species names follow Retief & Herman (1997). Soil forms are in accordance with Mac Vicar et al. (1991).

A Global Positioning System (GPS) reading in longitude and latitude, as well as a terrain description, were taken at each sample plot to facilitate future location ofthe specific vegetation unit. Habitat factors were also recorded and included the following: terrain type

(Land Type Survey Staff 1987; 1988; 1989), aspect, slope, topography, geology (Visser et

al.

1989), soil type (MacVicar et al. 1991), geomorphology, gradient, percentage rock cover, rock size and degree of erosion where applicable.

In each sample plot all species were recorded and a cover-abundance value was estimated with the Braun-Blanquet cover-abundance scale (Mueller-Dombois & Ellenberg

1974), namely r: rare;

+:

<1% cover; 1: <5% cover; 2a: 5-12% cover; 2b: 12-25% cover;

3: 25-50% cover; 4: 50-75% cover; 5: >75% cover. The vegetation structure at each releve was described according to a system of structural classification (Edwards 1983). All releve data are stored in the TURBOVEG database (Hennekens 1 996a) managed by the

Department of Botany, University of Pretoria (Mucina et al. 2000).

Endemic, near-endemic and Red Data List specieS/infraspecific taxa of the SCPE were determined from relevant literature (Hilton-Taylor 1996; Siebert 1998), fieldwork and herbarium surveys. The following symbols are used:

$

= endemic to the SCPE; #

= nearendemic to the SCPE; E = Endangered; V = Vulnerable; R = Rare; I = Indeterminate; K =

Insufficiently Known; N

= not threatened in northern provinces of South Africa (threatened

36

in one or more of the other provinces). New IUCN categories are still in the process of being applied or updated for most of these and other taxa (Golding 1999) and will be discussed elsewhere in this thesis.

3.1.2 Synthetic phase

A data set of 415 releves, containing a total of 1010 taxa, was entered into a vegetation database created in TURBOVEG (Hennekens 1996a). Two unpublished phytosociological data sets were available and included as part of the 415 releves, namely nine from a survey in the Maandagshoek region (Kritzinger 1992), and 44 from a study of the Potlake Nature

Reserve by MM Matthee during 1978.

As a first approximation the data was analysed with Two Way INdicator SPecies

ANalysis (TWINSP

AN) procedures (Hill 1979a) and based on the procedure developed by

Bredenkamp & Bezuidenhout (1995) for large data sets. TWINSPAN is a divisive, hierarchical classification technique which detects overall patterns of differences in biological data. Although the reliability of the TWINSP AN approach has been questioned under certain conditions (Van Groenewoud 1992; Van der MaareI1996), it was chosen for its proven combination of effectiveness, robustness and relative objectivity, as well as its availability and speed (Gauch

&

Whittaker 1981; Myklestad

&

Birks 1993).

To reduce distortion of data in the numerical data set, cut levels were adjusted in

MEGAT AB (Hennekens 1996b) to alter the default definition of pseudospecies, which insured less overweighing of dominants. A synoptic table was constructed to represent the major groups defined by the TWINSP AN classification (Table 1). Refinement of the synoptic table was done with Braun-Blanquet procedures according to the steps proposed by Behr

&

Bredenkamp (1988). A first step of an objective multivariate classification identified several vegetation types/major groups. The synoptic table contained species in each of the identified major groups on constancy values of a 20% ordinal scale (I-V). Only species with a minimum constancy value of 20%

(II), in any given major group, were included in the table.

37

This result was then used to subdivide the data set into five phytosociological tables, each representing one of the major vegetation types of the Sekhukhuneland study area

(Siebert

et at.

2002a). Each of these was again subjected to TWINSP AN. The resultant classification was further refined by using Braun Blanquet procedures in the MEGAT AB computer programme (Hennekens 1996b). The groups obtained from this data set were subsequently described and classified in Chapters 5 to 9.

The ordination algorithm DEtrended CORrespondence ANAlysis (DECORANA) (Hill

1979b) was applied to determine gradients in vegetation and the relationship between these plant communities and the physical environment. Results are depicted on a scatter diagram.

The ordinations are presented for the plant communities of each of the major vegetation types.

Plant communities were named binomially according to the recommendations of

Barkman

et at.

(1986). The first scientific name is that of a diagnostic plant species within the specific community. The second scientific name is that of a dominant species. Diagnostic and dominant species follow the definitions ofWerger (1974). An applicable physiognomic term is added to the community name in certain circumstances.

To facilitate the identification of areas of high conservation potential, the alpha diversities of the different plant communities were calculated. The alpha diversity (plant species richness) is defined as the number of species per unit area within a homogeneous community or the total number of species per community (Whittaker 1977). A 400 m

2 sample plot was taken as the unit area within a homogeneous community.

3.2

Plant and soil analyses

Analytical techniques to determine element concentrations in plants and soils have been used extensively in studies of ultramafic substrates and its associated vegetation. Examples of analytical approaches at the fine-scale level of heavy metal accumulation, speciation and mine rehabilitation in southern Afiica include studies by Wild (1974a), Brooks & Yang

38

(1984), Morrey et al. (1989), Hughes & Noble (1991), Roberts & Proctor (1992),

Anderson

e/ al.

(1995) and Balkwill & Burlin (1995).

As a result of a limited research budget and the high costs associated with mineral/metal analysis, soil and plant sampling was restricted to a few samples along a transect of a catena on the Critical Zone, in the undisturbed, under-utilised areas south of the Steelpoort River.

In addition, it was thought appropriate to only include a fixed set of heavy metals as part of the element spectrum investigated in both the plant and soil analyses of the SCPE.

Aluminium was selected, as it is the most common metal in the world's soils; it is a problem on 30-40% of the world's arable lands where acid soil releases aluminium ions into the ground water (Barinaga 1997). Chromium and nickel was chosen because the Critical Zone has the highest concentrations of Cr and Ni in the world (Schurmann et al. 1998) and these metals are typical for serpentines (Brooks 1987). Other parts of the SCPE is mined extensively for Fe, Mn and V, and because the ultramafic flora of Sekhukhuneland occurs on all four layers of the Rustenburg Layered Suite, it was decided to include these metals in the analysis.

3.2.1 Pot experiment

Topsoil and subsoil samples were collected from a chromium outcrop near Tweefontein,

Kennedy's Vale, Sekhukhuneland. Twenty-five samples were taken 500 mm deep in close proximity to established stands of sparse natural vegetation. Material was collected from the three dominant grass species All the samples were mixed to make up one soil mixture. The soil mixture was stored at room temperature in sealed bottles for 6 months.

Approximately 500 g of the soil mixture was weighed and placed in each of the 27 (9 x

3 sets) numbered containers and placed in a greenhouse at the beginning of spring 1999. A control soil of quartzite sand was also weighed into 27 (9 x 3 sets) numbered containers in the same greenhouse. A commercial fertilizer P:K:N (3:2:3), equivalent to a rate of application of 600 kg/ha (1 g per pot) was placed in the centre of each container below the surface. Maize seeds of the variety SNK 2340 (Vryburg), pre-treated with molybdenum, were planted in the containers at a depth of 10 mm to straddle the fertilizer band. Seeds were germinated and grown with rainwater. Plants were thinned to the required numbers

39

per container after 7 days. Even sized plants were selected for the experiment. The commercial fertilizer was re-applied every two weeks.

Different experimental layouts and procedures were used. Three different layouts were followed with regard to the number of plants left per container after thinning. The layouts had 2, 4 or 8 plants per container. Three different procedures were also followed with regards to harvesting. One pot of each layout was harvested after 2, 3 and 4 weeks respectively. On harvesting days the roots and leaves of plants were measured and then separately placed in well-marked paper bags. The harvested material was then dried in a drying oven for seven days. After drying, the dry mass was determined for both the roots and leaves. The dry material of plants grown in the chromium outcrop soils was milled and then analysed (see 3.2.3 Sample preparation and analyses) for a selected few elements, namely N, P, S, Mg, Ca, Ni and Cr.

The three most abundant grass species from the sites where the soil samples were taken were also sampled. These species were prepared and analysed according to the method in

3.2.3 Plant analyses. The element levels in these grasses were used as a control, as they belong to the same family as maize, namely the Poaceae.

3.2.2 Soil analyses

Rock samples

Rock samples were collected from 12 rock outcrops in randomly selected sites in the SCPE where vegetation anomalies occurred. A control rock sample of "true" serpentine was also collected from the Barberton Greenstone Belt in Mpumalanga. All the rocks were analysed and their metal and element concentrations determined. The area where the rocks were most strongly related to serpentine was chosen for a transect study.

Soil samples

Soil samples were collected along a transect of a catena near Tweefontein, Kennedy's Vale,

Sekhukhuneland. The mother material from this catena is of ultramafic origin. This site was proclaimed for Cr mining in 1999 and has subsequently been mined.

40

The catena was divided into 13 topographic positions (see 10.3.2 Results and discussion; Figure 24). Topsoil and subsoil samples were taken 500 mm deep in the rooting area of plant species that were in proximity to established stands of sparsely distributed natural vegetation anomalies (the collection of soil samples was used as an indication for the collection of plant material). Five soil samples were taken for each topographic position, close to the stems of the plant species that were sampled for analysis. Soil samples were stored in sealed bottles at room temperature for

6 months.

Sample preparation for both rock and soil analysis

Samples were dried and grinded to <75 IlIIl in a Tungsten Carbide milling vessel. Quartz crucibles were boiled in I: I HCL for 30 minutes in a glass beaker on a hot plate in a fume cupboard. Afterwards the crucibles were rinsed with distilled water and dried in a furnace at

I OOO°C for a 30 minutes and left to cool in a desiccator.

The cool, empty crucible was weighed on an analytical balance and its weight recorded to the fifth decimal place. Powder of the sample (2 g) was added and the weight recorded.

The crucible with the sample was dried at 110°C for a minimum of four hours to determine the amount of hygroscopic water in the sample. After the crucible was cooled in the desiccator, the sample was reweighed and the weight recorded. The sample was then ashed at a temperature of 950°C for four hours. After cooling in the desiccator, the sample was once again reweighed and the weight recorded. The Loss On Ignition (LOI) value is the sum of all changes that occur in a sample at a temperature of 950°C, expressed as a weight percentage of the original sample weight.

Fused beads were prepared following the standard method used in the XRD & XRF laboratory of the University of Pretoria (adapted from Bennett & Oliver 1992). A bead was prepared by adding I g pre-roasted sample to 6 g Lithium Tetra Borate (Li,B

4

0

7) flux and mixing by rolling it in a polytop covered with Mylar foil and a lid. Three drops of 250 gil

LiBr solution was added to a cleaned 5% AulPt crucible. The mixture was fused at

I

050

G

C in a muffle furnace with occasional swirling every 5 minutes. When no undissolved residue remained, it was poured into a pre-heated

Ptl

Au mould. in the furnace. The casting disk was

41

then removed from the furnace and placed on a refractory brick to cool. When the fused beads (glass disks) cooled down, they were removed from the casting dishes by tipping them upside down on paper.

Pellets were prepared using an adaptation of the method described by Watson (1996).

Using 16-20 mI, the grinded powdered samples were bound with 10-15 drops of a saturated movial solution. Everything was transferred into a plastic bag and rubbed thoroughly between the palms to distribute the movial binder evenly throughout the sample.

Samples were then transferred into an AI cup and pressed into a pellet under 8 tons/in

2 for two minutes. The pellet was then removed from the press and dried at 1 lOoC for 30 minutes.

Sample analysis

Rock and soil samples were analysed with X-Ray Fluorescence at the Department of

Geology, University of Pretoria. The apparatus used was the ARL 9400XP+ Wavelength

Dispersive X-Ray Fluorescence (XRF) Spectrometer. The XRF Spectrometer calculates the concentrations of the elements and a printout is obtained with all the results. The following method was followed:

I. Major element analysis was executed on fused beads, following the standard method used in the XRD & XRF laboratory of the University of Pretoria (adapted from Bennett & Oliver

1992). The bottom surface of a fused bead (glass disk) was analysed.

II.

Trace elements were analysed on pressed pellets, using an adaption of the method described by Watson (1996). Samples pressed into pellet under 8 tons/in

2 were analysed.

III. XRF Spectrometer was calibrated with certified reference materials. NBSGSC

Fundamental Parameter Programme was used for matrix correction of major elements as well as CI, Co, Cr, V, Ba, Sc and S. The Rh Compton peak ratio method was used for the other trace elements. Analyses were executed using the wide confidence limit program. The wide confidence limit program (QUANT AS) functions by executing a scan over the total wavelength span of the spectrometer using different crystaVwavelength combinations. The

42

overlap and background corrected peaks were quantified after application of the NBSGSC program for matrix correction.

There are two very different types of analysis that are used for the detennination of chromium:

(I) total chromium (with consideration to its oxidation state) and (2) chromium

(VI) (Kimborough

et af.

1999). Chromium (III) can be inferred from the difference between the two analyses. The analysis for total chromium is less complex and controversial than the analysis for chromium (VI). Since exchangeable levels of Cr are normally very low

(Kimbrough

et af.

1999), only total Cr was determined for the soil and plant material.

IV. Soil pH was measured in 1:2.5 soil:distilled water suspensions. The mixtures were left for one hour and measured with a pH metre at 25°C.

3.2.3

Plant

analyses

Plant material collection

Plant material was collected along a transect of a catena in the study area near Tweefontein,

Kennedy's Vale, Sekhukhuneland. Plant material samples were taken as roots, stems and leaves within established stands of natural vegetation. Certain criteria were followed for the selection of plant species:

• More than five specimens were available for collection;

• It is dominant in the plant community/vegetation anomaly;

• It is a SCPE endemic, near-endemic, form of a common species or disjunct locality.

Voucher specimens were prepared for each taxon sampled and are housed in the

H.G.W.J. Schweickerdt Herbarium (PRU), University of Pretoria. Collectors numbers are given in square brackets. This is followed by the reason why the specific species was cosen.

Plant material of the following plant species were collected:

Monocotyledons

Poaceae

Diheteropogon amplectens (Nees) Clayton [Siebert 671); dominant in community

43

Heteropogon contortus

(L.) Room.

&

Schult. [Siebert 600); dominant in community

Stipagrostis hirtigluma (Trin. & Rupr.) De Winter subsp. patula (Hack.) De

Dicotyledons

Winter [Siebert 597]; disjunct locality

Acanthaceae

Petalidium oblongifolium C.B. Clarke [Siebert 598); near-endemic species

Anacardiaceae

Rhus batophylla Codd [Siebert 936); endemic species

Rhus keetii Schonland [Siebert 931); near-endemic species

Asteraceae

Berkheya insignis (Harv.) Theil. [Siebert 942); endemic form

Brachylaena ilicifolia (Lam.) E. Phillips & Schweick. [Siebert 613); endemic form

Dicoma gerrardii Harv. ex F.C. Wilson [Siebert 929); dominant in community

Pterothrix spinescens DC. [Siebert 928); disjunct locality

Celastraceae

Catha transvaalensis Codd [Siebert 604); endemic species

Combretaceae

Terminalia prunioides G. Lawson [Siebert 605); dominant in community

Convolvulaceae

Ipomoea bathycolpos Hallier f. var. sinuatodentata Hallier f.

[Siebert 617]; endemic infra specific taxon

Ebenaceae

Euclea sp. nov.

(E.

sekhukhuniensis Siebert, Retief & Van Wyk) [Siebert 937]; endemic

Euclea linearis Zeyh. ex Hiem [Siebert 938); endemic form

Lamiaceae

Leucas capensis (Benth.) Eng\. [Siebert 596); endemic form

Orthosiphon fruticosus Codd [Siebert 615]; endemic species

Tinnea rhodesiana S. Moore [Siebert 614); dominant in community

Polygalaceae

Polygala sp. nov. (P. sekhukhuniensis Siebert, Retief

&

Van Wyk) [Siebert 602); endemic specIes

Schrophulariaceae

Jamesbrittenia aurantiaca (Burch.) Hilliard [Siebert 930); dominant in community

44

Plant material was collected along the broader topographic positions of the catena

(10.3.2 Results and discussion; Figure 24), namely chromium outcrops (I-M), the associated hill slope (E-I) and the eroded areas (A-E) in the valley below the slope. This collection method was followed because plant species follow the broader topographic trends. Plant material of succulents was not sampled, as Wild (1975) found that amongst endemics of Zimbabwean ultramafics, succulent species characteristically accumulate less heavy metals than non-succulent species. Monocotyledons growing on ultramafic soils often also have lower concentrations of metal ions in their tissue, largely because of preferential accumulations of elements in roots which are readily shed when the metal content becomes to high (Ernst 1972).

It was therefore thought best to concentrate all effort and expenses on non-succulent dicotyledons to ensure optimum results.

The dried plant material was stored in paper bags at room temperature for 6 months.

Before analysing the plant material it was milled to a fine powder. Plant material was analysed at the Institute for Soil, Climate and Water in Pretoria.

Sample preparation and analysis

L

Method for N (nitrogen) determination (Bel\omonte

et al. 1987).

The dried and milled sample was used directly for N determination on a Carlo Erba NA

1500 C/N/S Analyser (Dumas Method). A few milligrams of the sample was weighed into a tin container and ignited at high temperature in oxygen (on a chrome oxide catalyst). The gasses produced passed through silvered cobalt oxide, a column of copper (reducing the oxides of nitrogen to nitrogen gas and removing the excess O

2) and water vapour, and CO

2 traps. Gasses are then separated by gas chromatography using a helium carrier gas and detected by a thermal conductivity detector. The instrument was calibrated against a pure organic compound of known composition, in this case an ethyl ester of 4-Aminobenzoic acid, which contains 8,48% N.

II. Methods for the digestion and determination of Ca (calcium), Mg (magnesium), P

(phosphorous), S (sulphur), Fe (iron), Mn (manganese) and AI (aluminium).

45

Sample digestion (Zasoski & Burau 1977): Ig of a sample was digested with 7 mI

HN0

3

(concentrated nitric acid) and 3 mI HCl04 (perchloric acid) at a temperature of

200°C and brought to volume in a 100 ml volumetric flask.

Ca, Mg, Fe and Mn (Antanasopoulos undated): The solution was analysed with Flame

Atomic Absorption Spectrophotometry (AAS) for Fe and

Mn, using an Air-Acetylene

Flame with wavelengths of248.3 nm and 279.5 nm for Fe and Mn respectively. An aliquot of the solution was diluted for determination of Ca and Mg by AAS in a Nitrous Oxide-

Acetylene Flame, using a wavelength of 422.7 nm for Ca and 285.2 nm for Mg.

P and Al (Anonymous 1972; Hambleton 1990; AO.A.C. 1990): Other aliquots were used for the colourimetric determination ofP and AI, using automated flow systems. The P method uses the reaction of the phosphate with ammonium molybdovanadate and measurement of the absorption of the coloured complex at 420 nm. The Aluminon reagent method was used for AI (Jayman

&

Sivasubramaniam 1974; Bertsch

et al. 1981).

S (Ogner & Haugen 1977; Van Vliet 1999): A final aliquot was used for the determination of S (in the form of S04·

2

) by precipitation of barium sulphate, suspension of the precipitate in polyvinyl chloride and measurement of the turbidity.

III. Methods for the digestion and determination of Cr (chromium), Ni (nickel) and V

(vanadium).

Sample digestion (Chao-Yong & Schulte 1985): 15 mI HN0

3

(concentrated nitric acid) was added to a 0.5 g sample and heated to 120°C. After addition of 30% H

2

0

2

(10 drops) and a few ml distilled water, the samples were digested at this temperature for another 20 minutes, before cooling and bringing to volume in a 100ml volumetric flask.

Cr, Ni and V These 3 elements are simultaneously determined by ICP-MS (Inductively

Coupled Plasma-Mass Spectrometry). The isotopes used were V 51, Cr 52 and Ni 60. An internal standard (Indium - In) is used to increase the accuracy (added by diluting the digest solution with indium nitrate to 10 ppb. The procedures are based on the standard operating

46

concentrations and nutrient levels were once again determined by adding the totals of the applicable elements for each.

The third group of scatter diagrams relates element levels of the soil to that in the plant material. These diagrams aim to determine whether there is an association between the plant species and the soils on which they grow. This association is expressed as a function of the spatial distribution of plots along a two-dimensional plane (Figure 30). Figures used for the metal concentrations and nutrient levels were also determined by adding the totals of the applicable elements for each.

Plots of Cr and Ni concentrations of plant material against levels in the soil defined the fourth group of scatter diagrams (Figure 31). These are simple graphs indicating at what critical soil concentrations indigenous plant species accumulated these elements at maximum levels.

3.3 Floristic evaluation

Existing data on the distribution of plant taxa were obtained from PRECIS (National

Herbarium (PRE) Computerised Information System) (Prentice

&

Arnold 1997).

Distribution patterns of endemic plant taxa were projected by a Geographical Information

System [GIS], Idrisi for Windows 32® (Clark Labs 1999). This information was verified and supplemented by a study of herbarium specimens in the National Herbarium (PRE),

Pretoria, the HG.W.J. Schweickerdt Herbarium (PRU), University of Pretoria, and botanical literature.

Extensive fieldwork was conducted over two years in spnng and summer, and approximately 2 000 herbarium specimens were collected in the SCPE. The collections are housed in recognised herbaria of South Africa, namely the C.E. Moss Herbarium, University of the Witwatersrand (J), National Museum Herbarium, Bloemfontein (NMB), HG.W.J.

Schweickerdt Herbarium, University of Pretoria (PRU), and the National Herbarium,

Pretoria (PRE). More details on the composition of the checklist are supplied in Appendix

5.

48

Quantitative criteria were used to place a species in a particular Red List category using the guidelines set by the IUCN-Species Survival Commission (IUCN 1994). In most cases decisions were based on fieldwork observations and recorded locality, ecological and population data. Extent of occurrence was calculated for each species (Siebert 1998) using the IDRISI for Windows Geographic Information System package. Data were not analysed using RAMAS® Red List, a software package developed by a software development company, Applied Biomathematics (this software implements the IUCN Red List criteria for classifYing species into one of the three categories of threat, or a Low Risk category; if insufficient data is available to arrive at a conclusion it is classified as Data Deficient).

49

CHAPTER 4

PHYTOSOCIOLOGICAL STUDY

Abstract

A detailed account is given of the vegetation types of the Sekhukhuneland Centre of Plant

Endemism. Phytosociological data from 415 sample plots were subjected to phytosociological

classification using TWINS PAN The resulting classification was forther refined with table-sorting procedures based on the Braun-Blanquet floristic-sociological approach of vegetation classification. The analysis revealed six major vegetation types (Siebert

et al.

2002a) consisting of

82 syntaxa, interpreted as the

Acacia tortilis-Dichrostachys cinerea

Arid Northern Bushveld, the

Kirkia wilmsii-Tenninalia prunioides

Closed Mountain Bushveld, the

Combretum hereroense-

Grewia vemicosa

Open Mountain Bushveld, the

Hippobromus pauciflorus-Rhoicissus tridentata

Rock Outcrop Vegetation, the

Themeda triandra-Senecio microglossus

Cool Moist Grasslands and the

Fuirena pubescens-Schoenoplectus corymbosus

Wetland Vegetation. Plant communities of each major vegetation type are described and the diagnostic species highlighted. The occurrence of rare and threatened plant species in each plant community

is

indicated.

4.1 Introduction

The Sekhukhuneland Centre of Plant Endemism has a remarkably rich diversity of plant communities that are a direct result of the diversity of its substrates (geology and soils), climate (rainfall, temperature and fire patterns), topography (aspect, slope and height above sea level), floristic history (sub-tropical bushveld and afromontane ecotone) and human influence (agriculture, settlements, over-grazing and mining). These conspicuous plant communities constitute the various vegetation types of the centre, providing shelter and food for a myriad of organisms by means of tight, integrated local ecosystems. In addition these plant communities are of immense practical value to man and support the livelihoods of many of the rural people in Sekhukhuneland, both commercial and subsistence farmers

(Crooks

et al. 2000).

50

During the past few decades conservation management has moved in the direction of environmental management-the influence of human activities as they affect the quality of mankind's physical environment, especially air, water and terrestrial features (Sewell 1975).

Emphasis is not on strictly policed, protected areas primarily for large mammals, but on sustainable resource use, maintenance of ecological processes, and genetic diversity

(Cunningham 1989).

Savannas can be described as a tropical vegetation type co-dominated by woody plants and grasses. It is the dominant vegetation of Africa, occupying 54% of southern Africa

(Scholes 1997). The Savanna region is species-rich and 43% of the species are endemic to the subcontinent. This vegetation type is also home to many large mammal species (Cowling

& Olivier 1992). The central Savanna is an important location of biological diversity in the region and according to Rebelo (1997), 9.96% of the Savanna Biome is conserved in South

Africa. Furthermore, small land areas of only 7% in Mpumalanga and 2% in the Northern

Province has been set aside for conservation (Hoffinan

&

Ashwell 2000a; 2000b).

Many informal settlements exist in the central Savanna. These people rely on the savanna to supply grazing, fuel wood and timber. The SCPE lies in the savanna of the

Northern Province and Mpumalanga. Approximately 46% of Mpumalanga and 58% of the

Northern Province land areas is used for grazing (Hoffinan & Ashwell 2000a; 2000b). A further 30% and 22% of the land area respectively, is used for agriculture (Hoffinan

&

Ashwell 2000a; 2000b). The area contributes considerably to the formal economy of the region through its livestock, ecotourism and mining industries. Fast growing human populations in South Africa is making increasing demands on the natural resources and this will encourage expansion of agriculture and industry into marginal and often sensitive areas.

It is thus essential to have the necessary ecological knowledge of an area to assist in planning development, management and conservation to prevent future environmental deterioration.

Existing ecological knowledge of the vegetation types of the central Savanna of South

Africa is scanty (Cole 1986; Winterbach 1998) and confined to farms or nature reserves of local significance. Even less studied is the savanna of ultramafic substrates (norite,

51

anorthosite and pyroxenite) of the region (Siebert 1998). The first step to identifY broader vegetation types of the region, which also covered the ultramafic substrates of the western

Bushve1d Complex, was taken by Van der Meulen (1979).

This chapter deals with the analysis of ecological data to investigate the inten·e1ationships between plant communities and their environment (synecology). It also draws on other disciplines such as climatology, physical geography and pedology. This chapter provides an invaluable identification and classification reference to 82 of the common plant communities in the SCPE region, both indigenous and anthropogenic. A principal aim of the classifications presented in this chapter, is to define and describe the characteristics of the SCPE communities. This will assist scientists, conservationists and land-use planners when future projects are conducted in the centre. It is anticipated that the work presented win contribute to a more sustainable and less destructive development of the natural environment of the region. Sound environmental development is a state of mind and is something that can be achieved ifbasic data, such as this thesis, are actively drawn on during planning and management of natural resources.

The vegetation discussed here is largely confined to the norite and pyroxenite hins of

Sekhukhuneland. These substrates are intermediate between serpentine and granite-to quote Wild (1965) page 51, paragraph 4, on his view of the vegetation on such substrates:

"These intermediate characteristics of pyroxenite and norite soils, together with their

variability,

render an exact study of their vegetation more difficult than in the case of serpentine and so no attempt has been made to analyse their flora in detail here ... "

The heterogeneity of the vegetation on norites and pyroxenites was also recognised by

Acocks (1953) who described the Steelpoort area in Sekhukhuneland as a distinct variation of Mixed Bushve1d. The fact that he did not give it a distinguishing name at the time indicates the lack of knowledge that surrounds this vegetation type. This chapter is therefore an attempt towards classifYing the heterogeneous vegetation of the

Sekhukhune1and Centre of Plant Endemism.

52

4.2 Major vegetation types

Major vegetation types of the SCPE can be divided into three continuous regions, namely

Arid Bushveld, Mountain Bushveld and Grassland (Figure 6). These floristic regions are based on the broad distribution of major vegetation types. This floristic classification is hierarchical and dependent on scale, with smaller areas accommodated within successively larger ones (Maclaughlin 1992).

The first TWINSPAN division separated the azonal Wetland Vegetation from the other vegetation types. The second division separated the Arid Northern Bushveld from the moister southern and central vegetation types. A further division divided the vegetation into

Cool Moist Grassland and woodland/thicket vegetation types. A fourth division divided the bushveld into Rock Outcrop Vegetation, with afromontane elements, and mountain bushveld. Final division of the central mountain bushveld resulted in two types, namely

Open Mountain Bushveld and Closed Mountain Bushveld (Figure 33; Chapter 12).

Sekhukhuneland Centre of Plant Endemism

Fuirena pubescens-Schoenoplectus corymbosus

Wetland Vegetation

Acacia tottJlis-Dichrostachys cinerea

Arid Northern Bushveld

Themeda Iriandra-Senec/o microglossus

Cool Moist Grasslands

Hippobromus pauciflorus-Rhoicissus tridentals

Rock Outcrop Vegetation

Combrelum hereroense-Grewia vemicosa

Open Mountain Bushveld

Kirkia wilmsJl- Terminalla prunioldes

Closed Mountain Bushveld

Figure 33 Dendrogram depicting the TWINSPAN division of the six nwjor vegetation types of the Sekhukhuneland Centre of Plant Endemism (Dotted lines demarcate the vegetation types that are part of the proposed

Kirkia "lfilm!}il-Acacia ca/fra

Alliance on clay soils)

The most diagnostic species for each major vegetation type were distinguished, and based on the distribution of the plant species within the SCPE in general; the most

53

prominent character and differential species were used for the classification of the groups.

However, this remains provisional, for the vegetation of Sekhukhuneland is a complex system due to its heterogeneous habitats. It is difficult to predict the most prominent differential species, for significant variation in species composition arises in any given place and time. No two plant communities are identical in size, species composition or structure in the SCPE.

Endemic, near-endemic and Red Data List species/intraspecific taxa are given for each of the major vegetation types. Fifty-two endemic and 52 (of approximately 70) nearendemic species/intraspecific taxa (Siebert 1998) were recorded during the study. Thirtyseven taxa were identified as Red Data List taxa (Hilton-Taylor 1996), namely one

Endangered, two Vulnerable, eight Rare, one Indeterminate, 15 Insufficiently Known and

10 threatened in other provinces/countries of southern Afiica (not threatened in northern

Provinces).

The floristic composition of the six major vegetation types is given in the synoptic table

(Table 1). A discussion of the major groups follows below:

A.

Fuirena pubescens-Schoenoplectus corymbosus

Wetland Vegetation (Chapter 5)'

This wetland vegetation is found throughout the region, on stream banks in the valleys, seepage areas on the mountain slopes and wetlands on the mountain plateaus.

It is usually associated with vertic black clay soils that are saturated with water during the spring, summer and autumn seasons.

A floristic affinity exists with the Themeda triandra-Senecio microglossus Cool Moist

Grassland. It is also an extension of the wetlands on the Steenkampsberg (Bloem 1988).

This vegetation type is found throughout the Centre in all the floristic regions (Figure 6), especially in the grassland. It has, however, not been investigated thoroughly during this study.

IFor the purpose of describing the syntaxa, this major vegetation type is discussed with the grassland communities in Chapter 5.

54

This vegetation type is not bound by climate, geology, soils or topography, but is only dependent on a permanent water supply for the largest part of the year. Hence, many of the taxa in this major group are widespread throughout the northern provinces of South Africa.

Fuirena pubescens and Schoenoplectus corymbosus are the indicator species separating this azonal vegetation type from the zonal. Diagnostic species for this group are presented in species group R (Table 1).

Salix mucronata is the diagnostic woody species for the group. Herbs are plentiful, with

Artemisia afra, Conyza scabrida and Chironia purpurascens the diagnostic forbs and

Fimbristylis Jerruginea, Fuirena pubescens and Schoenoplectus corymbosus the diagnostic sedges. Frequently occurring, diagnostic taxa of the Poaceae include

Andropogon eucomis,

Imperata cylindrica, Miscanthus junceus and Phragmites australis, and other dominant graminoids are Cymbopogon validus and Hyparrhelliafilipendula.

This vegetation type has the lowest number of taxa of conservation value (Table 2).

However, this is a northeastern Drakensberg Escarpment wetland system, which means that it should receive conservation priority (Bloem 1988; Burgoyne 1995). An endemic form of

Acacia karroo, a near-endemic which is Insufficiently Known in the Red Data List, Nuxia

gracilis, and a Red Data List taxon not threatened in the northern provinces,

Eucomis

autumnalis subsp. clavata, occur in this vegetation type (Table 2).

B. Themeda triandra-Senecio microglossus Cool Moist Grasslands (Chapter 5)

This grassland is restricted to the higher altitude undulating hills of the southern region, and to a lesser degree, the high altitude plateau of the Leolo Mountains in the central region. It occurs on shallow clay soils underlain by norite and exhibits the highest floristic diversity in the region.

The vegetation is dense grassland, with scattered woody species. A floristic link exists with the grasslands of the Steenkampsberg (Burgoyne 1995). This major group is predominant in the Grassland floristic region (Figure 6).

55

High altitudes (Figure 2), temperate climates with high rainfall and frost (Figure 5), and seasonal fire gives rise to grasslands in the SCPE. This vegetation type follows the 600 mm and 18°C isohyet, and is maintained, not created, by the seasonal fires (Van Oudtshoorn

1999) that occur in different areas of Sekhukhuneland annually.

The most important indicator species for the division between the bushveld and the grassland are Diheteropogon amplectens and Senecio microglossus. Diagnostic taxa for this group are presented in species group L (Table 1).

Diagnostic woody species in this regIOn include the tree, Protea cajJra and the suffrutex, Elephantorrhiza elephantina. The invasive alien tree, Acacia dealbata, is a problem in this vegetation type. Many prominent forbs occur frequently in this major group and include the diagnostic Acalypha punctata, Clerodendrum triphyllum and Thesium

gracilentum,

and the abundant Berkheya insignis, Gnidia cajJra, Hypoxis rigidula, Senecio

lat!folius

and

S.

microglossus.

This vegetation type is characterised by the dominance of graminoids, which include prominent, conspicuous grasses such as Brachiaria serrata,

Diheteropogon amplectens, Elionurus muticus, Setaria sphacelata, Themeda triandra

and

Tristachya leucothrix.

The highest number of Red Data List taxa, namely 15, occurs in this vegetation type

(Table 2). Of these taxa two are Rare, seven are Insufficiently Known (highest number for the study area), one is Indeterminate and five are threatened elsewhere in southern Africa

(Table 2). This major group also has the highest number of taxa with conservation importance restricted to a vegetation type in the study area (15), and includes taxa such as the endemic Zantedeschia jucunda and the Rare Eucomis montana (Table 2).

C.

Hippobromus pauciflorus-Rhoicissus tridentata

Rock Outcrop Vegetation (Chapter 6)

The communities of the Hippobromus pauc!florus-Rhoicissus tridentata Rock Outcrop

Vegetation are scattered as bush clumps, or stages of it, throughout the study area, but are more frequent in the southern region. It prefers sheltered habitats of rock outcrops, classified as rocky outcrops, -ridges, -flats and -refugia. On a macro scale, the vegetation of rocky outcrops is dependent on topography (Fi!,'ure 2). However, this vegetation type,

56

although not diverse, is very specialised and a direct consequence of specific environmental conditions (Bredenkamp

&

Deutschlander 1995).

The vegetation type can be found within all the floristic regions of the Centre, but to a lesser degree in the Mixed Bushveld floristic region (Figure 6). These broad-leaved closed woodlands or open shrublands of rock outcrops have a strong floristic link with afromontane vegetation. Two patches of afromontane forests, both from the Leolo

Mountains, are included in this group. The forest tree layer is mostly 5 m, but heights of up to 10 m have also been recorded. These afromontane forests of the SCPE are undersampled and are provisionally grouped here until further research can provide more information that will probably lift the forest communities out ofthis major vegetation type into its own.

The indicator species that delimitate this vegetation type are Celtis africana and Aloe

arborescens. Diagnostic species of this Rock Outcrop Vegetation type are listed in species group G (Table 1).

Prominent tree/shrub species, representative of all four types of rock habitats are the diagnostic May tenus undata and the woody Acacia ataxacantha, Aloe castanea,

Cambre tum molle, Cussonia transvaalensis, Hippobromus pauciflorus and Rhoicissus

tridentata. The most abundant forbs include the diagnostic taxa Cyphostemma woodii,

Gerbera jamesonii, Orthosiphon labiatus and Tetradenia brevispicata. Xerophyta

retinervis is also prominent in the group. Dominant grasses are the diagnostic Aristida

transvaalensis and abundant Cymbopogon excavatus.

This vegetation type has the status as the major group with the highest number of SCPE near-endemic taxa (Table 2). The second highest number of Red Data List taxa is also present, including one of the two Indeterminate taxa recorded for the study area,

Aloe

reitzii var. reitzii (Table 2). Fifteen taxa of conservation importance, the second highest number for the SCPE, are restricted to this group, of which

Adenia wilmsii, Euphorbia

sekhukhuniensis and Tulbaghia coddii are of conservation priority (Table 2).

57

D. Combretum hereroense---Grewia vemicosa Open Mountain Bushveld (Chapter 7)

This sparse open bushveld has a patchy distribution throughout the whole study area. It occurs on anomalous soils that contain high concentrations of heavy metals (Al, Cr, Fe, Ni,

Pt, Ti and V) and high levels of Mg and Ca (see Chapter 10). These soils have a weak structure and high erosion potential.

This sparse bushveld, with a scattered grass sward, gives way to the

Acacia tortilis-

Dichrostachys cinerea Arid Northern Bushveld (a deciduous microphyllous thornveld) in the north and Kirkia wilmsii-Terminalia prunioides Closed Mountain Bushveld (a deciduous broad-leaved savanna) in the central parts where the soils are 'normal'. To a lesser extent it also occurs as patches in the

Themeda triandra-Senecio microglossus Cool

Moist Grasslands. Thus an extensive mosaic is formed. It is, however, more predominant in the Mountain Bushveld floristic region (Figure 6).

The existence of this vegetation type can primarily be ascribed to geology (Figure 3) and soils (Figure 4). Aridity, induced by freely drained or vertic soils, and metalliferous soils, produced by specific layers of the Rustenburg Layered Suite, have created harsh environments. These open niches have been filled by a specific group of plant species, which are common in other major groups as well. This vegetation type can be described as an anomaly, for the species composition and predominantly stunted structure is very distinctive and different from the surrounding vegetation.

Combretum hereroense and Loudetia simplex were identified as the indicator species that separate this vegetation type from the other bushveld types. Diagnostic plant species for this vegetation type are listed in species group D (Table 1).

Small trees/shrubs, which are diagnostic, are

Brachylaena ilicifolia and Ozoroa

sphaerocarpa, prominent and abundant woody species include Combretum hereroense,

Grewia vemicosa, Tinnea rhodesiana and Vitex obovata subsp. wilmsii. Forbs such as the diagnostic Euphorbia enormis and Orthosiphon fruticosus, and prominent Commelina

africana, Kyphocarpa angustifolia and Phyllanthus glaucophyllus are occurs frequently.

58

Enneapogon scoparius, Heteropogon contortus and Themeda triandra are the dominant grasses of the vegetation type.

This is the major vegetation type with the most SCPE endemics recorded within its plant communities (Table 2). Together with the Rock Outcrop Vegetation it is host to four

Rare taxa. The only Endangered taxon in the study area, Euphorbia barnardii, occurs in this vegetation type (Table 2).

E.

Kirkia wilmsii-Terrninalia prunioides

Closed Mountain Bushveld (Chapter 8)

This bushveld vegetation type occurs predominantly in the central parts of the SCPE, on clay soils of mountain slopes that are underlain by norite and pyroxenite. The topography is predominantly and typically an undulating landscape. It is the dominant group of the

Mountain Bushveld floristic region (Figure 6).

The grass layer of this woodland is well developed and the tree layer varies between 2-

5 m. A mosaic is formed with the Acacia tortilis-Dichrostachys cinerea Arid Northern

Bushveld on clays of the dry valleys and the

Combretum hereroense-Grewia vemicosa

Open Mountain Bushveld on anomalous soils of mountain foot slopes and valleys.

This major group is a product of the regions topography (Figure 2), soils (Figure 4) and climate (Figure 5). The relatively drier, warmer climate facilitated the development of bushveld instead of grassland on the hills.

An undulating topography separates this bushveld group from the lowland microphyllous thornveld of the Springbok Flats along the western border of the SCPE. Soil patterns were responsible for the division between Closed

Mountain Bushveld on 'normal' soils and Open Mountain Bushveld on 'toxic' soils.

The indicator species for this vegetation type are Dichrostachys cinerea and Panicum

deustum. These species are the most important taxa in the group's separation from the other related major groups. Plant species of diagnostic value in this vegetation type are listed in species group B (Table 1).

59

The

Kirkia wilmsii-Terminalia prunioides Closed Mountain Bushveld is characterised by the diagnostic trees

Acacia nigrescens and Commiphora mollis, and the dominant trees/shrubs Acacia senegal var. leiorachis, Combretum apiculatum, Kirkia wilmsii and

Terminalia prunioides. Conspicuous dominant forbs are the diagnostic Clerodendrum

ternatum, and the prominent Barleria saxatilis, Psiadia punctulata and Sanseviera

hyacinthoides. Prominent, abundant grass species include Aristida canescens, Enneapogon

scoparius, Heteropogon contortus and Panicum deustum.

This major group has the second highest number of SCPE endemics and SCPE nearendemics (Table 2). Ten Red Data List taxa were also recorded, with three taxa categorised as Rare, of which two are endemic to the SCPE (Table 2). Fourteen taxa of conservation importance are restricted to this group, of which the endemic,

Ledebouria dolomiticola, and two undescribed species are examples (Table 2).

F.

Acacia tortilis-Dichrostachys cinerea

Arid Northern Bushveld (Chapter 9)

This vegetation type occurs mostly in the moderately arid and warmer northern part of the study area. It is usually restricted to the deep, clayey alluvium soils of the Olifants and

Steelpoort River valleys in the Mixed Bushveld floristic region of the SCPE (Figure 6). It also occurs in the dry river valleys between the mountains of the central parts of the SCPE, where it forms a mosaic with the Kirkia wilmsii-Terminalia prunioides Closed Mountain

Bushveld.

This bushveld is characteristically a sparse thornveld with an open grassy layer. The tree layer usually reaches a height of approximately 3 m. A floristic relationship exists with the vegetation of the Pietersburg Plateau (Bredenkamp

&

Van Vuuren 1977).

This major vegetation type is climatically induced, more specifically by rainfall, for it is restricted to the region with a maximum average annual rainfall of 400 mm (Figure 5). The geology of this region is very heterogeneous (Figure 3) and the soils extremely diverse

(Figure 4), and are responsible for heterogeneity within communities.

60

The diagnostic graminoids Eragrostis barbinodis and Tragus berteronianus are the most important indicator species at the division level separating this bushveld vegetation type from the moister

Kirkia wilmsii-Terminalia prunioides and the Combretum

hereroense-Grewia vernicosa Mountain Bushveld types. All diagnostic species of this group are given in species group A (Table 1).

Acacia tortilis, Boscia foetida and Dichrostachys cinerea are the most abundant and prominent dominant tree species of the group. The most frequent occurring diagnostic forbs are Becium filamentosum, Felicia clavipilosa, Gisekia africana, Hermannia odorata and

Melhania rehmannii. Prominent graminoids of the vegetation type include Aristida

congesta, Enneapogon cenchroides,

E.

scoparius and Urochloa mossambicensis.

One Rare taxon,

Boscia foetida subsp. minima, and the only recorded Vulnerable taxon for the study area, Plinthus rehmannii, occur in this vegetation type (Table 2). Five taxa of conservation value are restricted to this vegetation type and include the succulent SCPE near-endemic,

Huernia stapelioides (Table 2). This major group has the lowest number of taxa with conservation value of all the zonal vegetation types.

4.3 Hierarchical classification

A hierarchical classification of all the syntaxa identified within the six major vegetation types is presented to give a clearer overview of the results. Each syntaxon is also linked to the page number on which it is discussed as part of relevant groups.

ISYNTAXON

I

PAGE

I

Themeda triandra-Senecio microglossus

Cool Moist Grasslands

L

Tristachya leucothrix- Trachypogon spicatus

class of mountain slopes and plateaus

\. Helichryso splendidi-Tristachyetum leucothricis

2. Zanfedeschio pentlandi-Aloetum castaneae

3. Brachiario serratae-Melhanietum randii

3.1 Brachiario serratae-Melhanietum randi; helichrysetosum rugu/osii

81

81

81

82

83

84

61

3.1.1 Digitaria eriantha variant

3.1.2 Alloteropsis semialala variant

3.2 Brachiario serralae-Melhanielum randii argyrolobielosum transvaalense

3.2.1 Koeleria capensis variant

3.2.2 Berkheya seminivea variant

3.3 Brachiario serratae-Melhanietum randii gnidiefosum capitatae

3.4 Brachiario serratae-Melhanietum randii setarietosum nigrirosh"s

4. Elionuro muticusae-Trachypogonetum spicati

4.1 Elionuro mulicusae-Trachypogonelum spicali bewsielosum bijlorae

4.2 Elionuro muticusae-Trachypogonetum spicati acacietosum tortilis

5. Jamesbrittenio macranthae-Loudetietum simplicis

5.1 Jamesbrittenio macranthae-Loudeh"etum simplicis combretetosum hereroense

5.2 Jameshrittenio macranthae-Loudetietum simplicis eucleetosum Iinearis

II. Fuirena pubescens-Schoenoplectus corymbosus

community of streams and seepage areas

6. Fuireno pubescenlis-Schoenietum nigricanlis

6.1 Fuireno pubescenlis-Schoenielum nigricanlis lriraphielosum andropogonoidis

6.2 Fuireno pubescentis-Schoenietum nigricantis pycnostachetosum reticula/ae

6.3 Fuireno pubescentis--Schoenietum nigricantis bulbostylietosum hispidulae

7. Andropogono eucomis-Fimbristyletumferrugineae

8. Limosello maioris-Ranunculetum meyeri

97

97

98

99

94

95

95

96

90

91

92

93

88

89

90

85

85

86

87

87

Hippohromus pauciJlorus-Rhoicissus tridentata

Rock Outcrop Vegetation

L

Rhoicissus sekhukhuniensis-FlCus abutilifoiUl

community of rocky outcrops l.

Vepro rejlexae-Mimusopelum zeyheri

2. Commiphoro marlolhii-Crolonelum gralissimi

IL

Cymbopogon excavatus-Pavetta

sp. nov. community of rocky ridges

3. Grewio monticolae-Elephantorrhizetum praetermissae

4. Melino nerviglumis-Calhelum edulis

5. Heleropogono conlorli-Apodylelum dimidialae

6. Gerbero jamesonii-Kirkietum wilmsii

7. Brachiario serratae-Viticetum wilmsi;

8. Cymbopogono excavali-Brachylaenelum rolundatae

9. Aloo pretoriensis-Xerophylelum retinervis

10. Tephrosio purpureae-Rhoicissetum /ridenlalae

11.

Cymbopogono validi-Rhamnelum prinoidis

12. Enleropogono macroslachis-Hippobromelum paucijlorii

120

120

121

122

123

123

124

125

126

127

128

128

129

130

131

62

m

Crassula sarcocaulis-Aristida transvaaIensis

commuuity of rocky flats

13. Munduleo sericeae-Euphorbirtum cooperi

14. Crassulo sarcocaulis-Aristidietum transvaalensis

IV.

Panicum deustum-Celtis africana

community of rocky refugia

15. Clauseno anisatae-Diospyretum whyteanae

16. Fico sur-Combretetum erythrophyllii

17. Andrachno ovalis-Allophylletum transvaalensis

Combre/um hereroense--Grewia vernicosa

Open Mountain Bushveld

I. Enneapogon scoparius-Comhretum molle

community of mountain slopes

1. Enteropogono macrostachyo-Sc/erocaryetum hirreae

1.1

Enteropogono macrostachyo-Sclerocaryetum birreae asparagetosum sekukuniensis

1.2

Enteropogono macrostachyo-Sclerocaryetum birreae grewietosum vernicosae

2. Enneapogono scoparii-Acacietum leiorachis

2.1 Enneapogono scoparii-Acacietum ieiorachis chloreiosum virgatae

2.2 Enneapogono scoparU-Acacietum /eiorachis grewietosum jlavescentis

2.3 Enneapogono scoparii-Acaciefum leiorachis brachylaenetosum i/icijoliae

2.4 Enneapogono scoparii-Acacietum leiorachis commiphoretosum mollis

3. Phyllantho glaucophyllae-Brachylaenetum ilicifoli

3.1 Phyllantho glaucophyllae-Brachylaenetum ilicifoli setarietosum sphacelatae

3.2 Phyllantho glaucophyllae-Brachylaenetum ilicifoli brachiarietosum serratae

4. Tristachyo leucothricis-Cussonietum transvaalensis

4.1 Tristachyo leucothricis-Cussonietum transvaalensis myrothamnetosumjlabellifolius

4.2 Tristachyo leueothricis-Cussonietum transvaalensis melinetosum nerviglumis

4.3 Tristachyo leucothricis-Cussonietum transvaalensis argylobietosum wilmsii

4.4 Tristaehyo leucothricis-Cussonietum transvaalensis eombretetosum zeyheri

II.

Loudetia simplex-Combretum hereroense

community of valleys

5. Eragrosti lehmannianae-Hippobrometum paucijlori

5.1 Eragrostio lehmannianae-Hippobrometum paucijlori rhoetosum batophyllae

5.2 Eragrosti lehmannianae-Hippobrometum paucijlori sorgetosum bieoloris

5.3 Eragrosti lehmannianae-Hippobrometum paucijlori elionuretosum mutici

6. Aristido rhiniochloo-Gnidietum polycephalae

7. Loudetio simplicis-Eucleetum linearis

7.1 Loudetio simplicis-Eucleetum linearis diheteropogonetosum amplectentis

7.2 Loudetio simplicis-Eucleetum linearis heteropogonetosum contorti

7.3 Loudetio simplicis-Eucleetum linearis andropogonetosum chinensis

63

158

163

164

165

166

167

168

169

169

170

158

159

160

161

162

171

172

173

174

175

176

177

178

178

179

180

181

182

132

132

133

134

135

136

136

8. Petalidio oblongi{olii-Raphionacmetum procumbentis

Kirkia

wilmsi~Terminalia

prunioides

Closed Mountain Bushveld

L

Enneapogon scoparius-Kirkia wilmsii

community of mountain slopes

I.

Combreto apiculati-Kirkietum wilmsii

1.1 Combreto apiculati-Kirkietum wilmsii clerodendretosum glabrae

1.2 Combreto apiculati-Kirkietum wilmsii eustachetosum paspaloidis

1.3 Combreto apiculati-Kirkietum wilmsii bridelietosum moWs

1.4 Comhreto apicu/ati-Kirkietum wilmsii chaetacanthetosum cos/alii

1.5 Combreto apiculati-Kirkietum -wilmsii hermannietosum boraginijlorae

1.6 Combreto apiculaii-Kirkietum wilmsii themedetosum triandrea

1.7 Combreto apiculati-Kirkietum wilmsii nuxietosum congestea

2. Panico deustii-Dichrostachetum cinereae

2.1 Panico deustii-Dichrostachetum cinereae sporoboletosum stapjianii

2.2 Panico deustii-Dichrostachetum cinereae maeruetosum ango/ensis

2.3 Panico deustii-Dichrostachetum cinereae melhanietosum prostratae

2.4 Panico deustii-Dichrostachetum cinereae melhanietosum acuminatae

II.

Eragrostis curvula-Combretum hereroense

community of valleys

3. Flngerhuthio africanae-Boscietum foetidae

3.1 Fingerhuthio africanae-Boscietum foetidae elaeodendretosum transvaalensis

3.2 Fingerhuthio africanae-Boscietum foetidae aloetosum giobuligemmae

3.3 Fingerhuthio africanae-Boscietum foetidae euphorbietosum ingentis

3.4 Fingerhuthio africanae-Boscietum foetidae sesamothamnetosum lugardii

4. Hippocrateo longipetiolatae-Euphorbietum tirucal/i

4.1 Hippocrateo /ongipetiolatae-Euphorbietum tirucalli emilietosum transvaalensis

4.2 Hippocrateo longipetiolatae-Euphorbietum tirucaJli aristidetosum transvaalensis

4.3 Hippocrateo longipetiolatae-Euphorbietum tirucalli bothriochloetosum insculptae

5. Celtido africanea-Combretetum erythrophyllii

5.1 Celtido africanea-Combretetum erythrophy/lii acacietosum caffrae

5.2 Celtido africanea-Combretetum erythrophyllii acacietosum galpinii

Acacia tortilis-Dichrostachys cinerea

Arid Northern Bushveld

L

Eragrostis barbinodis-Acacia tortilis

community of arid systems

1. Panico colorati-Crotonetum menyhartii

2.lvfe/hanio rehmannii-Acacietum torti/is

2.1 Melhanio rehmannii-Acacietum tortilis grewetosuma bieoloris

64

207

213

214

215

216

217

218

219

220

221

229

230

231

232

233

221

222

223

225

226

227

228

207

208

209

210

211

212

255

255

256

257

258

2.2 Me/hanio rehmannii-Acacietum tortilis rhigozetosum obovati

2.3 .lWe/hanio rehmannii-Acacietum torti/is diospyretosum /ycioidis

2.4 Afe/hanio rehmannii-Acacietum tortilis acacietosum niloticae

2.5lvfe/hanio rehmannii-Acacietum tortilis indigo!eretosum rhytidocarpae

3. Enneapogono cenchroidis-Salvadoretum australis

4. [kochloo panicoidis-Agavetum americanae

259

260

261

262

263

264

65

__

I

~

A i

..

Jane Furse

~

• I

l

I

I

1"

--

, ,

l

~ ##~

,/:; i

~

B

I

I

I

I

I

I

I

.:

R;ssenekai

CJ

/

(\

tl·

Burgerslort

,

f •

,-..

,-,

B :

,

,

\

.,

,

....

'

.,

.......

V

I

I

,

Lydehburge

"'--_.I .. ___ i /

~fI'#4"

~

STEENKAMlfSSERG

~

o

1 : : '

20II1II

====='

FLORISTIC REGION

A

=

ARID BUSHVELD

B

=

MOUNTAIN BUSHVELD

C

=

GRASSLAND

Figure

6 Major floristic regions identified for the Sekhukhuneland Centre of Plant Endemism

.

(Mountain Bushveld comprises Open and Closed Mountain Bushveld

;

Rocky Outcrop and

Wetland Vegetation is scattered throughout the stud

y

area)

.

66

0\

- l

Table 1 Synoptic table of the major vegetation types of the Sekhukhuneland Centre of Plant Endemism.

MAJOR GROUP

NUMBER OF RELEVES

1

2

3 4 5

6

47 103 91 100 57

17

MAJOR GROUP

NUMBER OF RELEVES

1 2

3 4 5 6

47 103 91 100 57 17

SPECIES GROUP A

Diagnostic for the

Olchrostschys clneres

Eragrostis barb/nod/s

~.

Tragus hefferon/anus

IV

.

Bec/um fllamentosum

III

I

Felicia clav/pilos.

Gisekia african.

Hermann/a odorata

Me/han/a rehmannii

Phyllanthus maderaspatensis

Urochloa mosambicensis

Acacia mellifenJ

Arist/da scabriva/Vi$

Blepharis integrlfolla

Hermann;a modest.

Hibiscus praeteritus

Indigastrum cost.tum

PfaeroxyJon obIiquum

Solanum coccineum

Acacia grandicomuta

Albizia anthelminaea

Cenchrus a/liarls

Corchorus asplenifolius

Indigofers enormis

Pfycholobium contortum

Schmidtill pappophoroides

III

III

III

III

III

II

I

I

I

I

I

I

I

I

I

I

I

I

I

I

Sporobo/us ioe/lldos

I

~

I

SPECIES GROUP B

Diagnostic for the

Klrki. wI/msll-Tennlns"a pronloldes

C/erodendrum tematum

Commiphora mol/is

I f!l

I

I

I

I

I

I

SPECIES GROUP B (coni.)

!l

Acacia nigrescerla

Acaclll nilotic_

~rloIfda

meridlana//o

ArloIfda

rh/nloch/oa

Sarlerla l.ncIfoI/a

Carissa

b/$pInOSIJ

Commiphore

african.

Cryptot.pis abIongifo/ium

DoIichoe

trIlobua

EucIH dlv/norum

Fluegge. vlroea

Grew/a nllVescen.

Indigchra Iydenburgensis

Oehns lnermi.

Ocimum canum

Rhynchosi. minima

Sansevierla hyacinthoidN

Sldadragei

Sporobolus flmbriatus

Sporobolus stapfianus

Sten:uJia rogers/I

Urginft

epigea

Ximen/a americana

SPECIES GROUP C

AClJCill torti/is

Sedde,." suffruticosll

Enneapogon cenchroides

Lantana rugosa

Monechma divaricatum

Boscia afbitrunca

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

~

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

I

IV II

I

IV II

III

II

III II

III

II

I

II

III I

I

I

I

I I

I

0-,

00

Table 1 continued.

MAJOR GROUP

NUMBER OF RELEVES

1 2 3

47

103 91

4 5

6

100 57 17

SPECIES GROUP C (conI.)

Grewis flavs

Boscia foetida

Cadaba tennitaria

Commiphora pyracsnthoides

Croton menyharlii

Digitalia eriantha

Ehret/a rigida

Eragrostis rigidior

Rhus englen"

II

II

II

II

II

II

III I

II

II

I

I

I

I

I

I

I

I

I

SPECIES GROUP D

Diagnostic for the

Combretum hereroense-Grewia

vem~a

SrlJchylaena ilicifolia

Orthosiphon fruticosus

Ozoroa

sphaerocarpa

Andropogon chinensis

Argyrolobium wUmsii

Bolusanthus speciosus

Casslne aethlopica

Catha edulis

Crabbe. angustifo/;a

EJaeodendron transvaalensis

Euclea linearls

.

I

I

I

I

I

I 111

I

I

I

III

III

II

II

II

II

II

II

II

II

I

I

I

I

I

I

Euclea

sp. (S 934)

Euphorbia enormis

Euphot'bia $chinzii

/ndigofera nebrowniana

I

I

II

II

II

II

I

Ipomoea bathyco/pos var. sinuatodentata

Jamesbrittenla atropurpurea

II

II

Laggera decurrens

Ledebouria marginata

I

II

II

I

I

I

I

I

MAJOR GROUP

NUMBER OF RELEVES

SPECIES GROUP D (con1.)

.. Ormocarpum kirkii

Po/wafs

sp.

(5 449)

Rhus batophylla

Rhuskeetii

Tarchonanfhus camphoratus

SPECIES GROUP E

Psiadia punctulata

Acacia senegal

var.

leiorachls

Aristlda canescens

Barlen. slJxatiJis

Combretum apicu/atum

Combretum hereroense

Grewia vemicosa

Tinn" rhodesiana

Phyl1anthus glaucophyllus

Aloe burgersforlensis

Chastacanfhus costatus

Dalechampia galpinii

Decorsea galpinii

Eragrostis lehmanniana

Fingerhuthia africana

Jatropha /stifolia

Just/cia protracta

Sclerocarya birrea

Waltheria indica

SPECIES GROUP F

Dichrostachys cinerea

Enneapogon scoparius

1

2

3 4 5 6

47 103 91

100 57 17

I

I

I

I

II

II

II

II

II

'--

I

I

I

I

I

I

I

I

I

I

I

IV II

III

III

I

II

II

II

III II

III II

III II

I II

IV

IV

IV

III

II

II

II

II

II

II

II

II

II

II

I

I

I

I

I

I

I

I

·1

I

I

I

I

IV IV II

III IV III

I

0-,

'"

Table 1 continued.

MAJOR GROUP

NUMBER OF RELEVES

1 2 3 4 5

47

103 91 100 57

6

17

SPECIES GROUP F (cont.)

Evo/vu/us alsino/des

Leucas capensis

Kyphocarps angustffolia

Aptosimum lineare

Balanites maughamii

S/epheris subvolubilis

Corbichonia decumbens

Geigeria amative

Petslidium ob/ongifO/ium

III

III

II

II

II

II

II III

II

II

II

II

II

II

I

I

I

SPECIES GROUP G

Diagnostic for the

Hlppobromus pauclflorus-Rholclssus

trid~ts

Aristida transvBalensis

CyphO$temma woodii

Gerbera jameson;;

.

I

I

I

I

I

III

III

III

III

Msytenus undsta

Allophylus africanus

Apodytes dimidiata

Asparagus intricatus

Serchemia zeyheri

Clematis brachiata

Crassula sareocaulis

Cyphia elata

Diospyros whyteana

I

I

I

I

I

I

II

I

I

I

I

I

I

DovyaJis zeyheri

Drimiopsis maxima

I

I

Grewia occidenta/is

HalJeria lucida

Olea capensis

Olin;a emarpinata

Orthosiphon labiatus

I

I

MAJOR GROUP

NUMBER OF RELEVES

SPECIES GROUP G (cont.)

Ruellia stenophylla

Sco/opia zeyheri

Tetradenia brevispicata

Vangueria infsusta

Zanfedeschia pentlandii

Zanthoxylum fhomcroffii

SPECIES GROUP H

Diospyros /ycioides

subsp. nitens

Acacia stsxacsntha

Combretum mo/le

Cymbopogon eXCBvatus

Xerophyte retinervis

Catha transvasfensis

Mimusops zeyheri

Pavetta zeyheri

Sphedamnocarpus pruriens

Thesium burke;

SPECIES GROUP I

Kirkia wrlmsii

Panicum deustum

Asparagus buchanan;;

EJephantorrhiza praetermissa

Hippobromus paucifTOIUs

Ziziphus mucronata

Aloe cryptopoda

Croton gratissimus

Dombeya rofundifolia

Jasminum multiparlitum

Kleinia stapeNiformis

1

2

3

4

5 6

47 103 91

100 57

17

I

I

I

I

I

" -

I

I

I

I

I

I

I

I

II

II

II

II

III

III

III

III

II

II

II

II

II

I

I

I

I

I

IV III

II

IV II

III

III

I

II

I

II

II

IV

I

I

I

I

I

I

I

II

-.l o

Table 1 continued.

MAJOR GROUP

NUMBER OF RELEVES

SPECIES GROUP I

(conI.)

Setaria Jindenbergiana

Triaspis g/aucophyJ/a

SPECIES GROUP

J

Aloe marlothii

Celtis africsna

Clerodendrum g/abrum

Diospyros lye/aides

subsp. lye/aides

Euphorbia ingens

Grew;a monticoia

Maytenus heterophy/Ja

Opuntia ficus-indica

Pappea capensis

Peltophorum africanum

Styfochiton nataJensis

1

2

3

47 103 91

4 5

100 57

6

17

II

II

II

II

II

II I

I

I

I r-rr-

I

I

II

I-,,-

II

II

II

II I

I

I

II

II

II

II

...!!...

I

I

I

I

~

SPECIES GROUP K

Kleinia longiflora

Sarcostemma viminale

Dodonaea angustifolia

Ka/anchoe pan/culata

Mundulea seric8a

Rhus gueinzii

Schotia bnJchypetB/a

Solanum panduriforme

IV II

I III

II

II

II

II

II

I

II

II

I

II

III

II

II

II

I

I

SPECIES GROUP L

Diagnostic for the Themeda rlBndra-Seneclo mlcroglossus

Acalypha punctata

Clerodendrum triphyllum

Thesium gracilentum

I

I-rri

III

III

MAJOR GROUP

NUMBER OF RELEVES

+-

SPECIES GROUP L

(conI.)

Athrixia s/ata

Ca/lilepis /eptophylla

Cephalari. zeyheriana

Dicoma zeyheri

Elephantotrhiza e/ephantinB

Eragrostis superba

Gnldia capitats

Hermannia anlonii

Me/han;a randii

Phyllanthus parvulus

Polyps/a uncinata

Prates

caffra

Rhoicissus

sp. (5 48)

Rhus wilms;;

Rhynchosia sordida

Scilla natalensis

Strlga asiatica

Sfriga bilabiats

Tetrase/ago wilmsii

Trachypogon spicatus

Tristachya rehmannH

Vernonia oIigocephaJa

SPECIES GROUP M

Hypoxis rlgidulB

Aloe greatheadii

Eragrostis nindensis

Ledebouria revoluta

Convolvulus sagittatus

Eragrostis pseudosc/erBntha

Eragrostis rBcemOSB

1 2 3

4

5 6

47 103 91

100 57

17

I

I I

I

I

I

I

I

I

I

I

I

I

I

I I

I

I

I

I

I

I

I

I

~

I

I

I

I

I

I

I

I

I

I

I

I

V

III

III

III

II

II

II

I

Table 1 continued.

MAJOR GROUP

NUMBER OF RELEVES

SPECIES GROUP M (cont.)

Jasminum quina tum

Lopholaena coriifO/is

Melinis repens

CyphootemmB

sp. A (W 13389)

Pearson;a sess/lifO/la

Senecio macrocephalus

SPECIES GROUP N

Vitex abovef.

subsp.

witmsN

Euclea crisps

Rhoicissus trident.t.

Cusson/a transvaa/ens;s

Senecio latifolius

Rhynchosia spectabilis

Setaria sphace/at.

Acacia csffra

Eragrostis chlcxomelas

Tristachya /eucothrix

SPECIES GROUP 0

Diheteropogon amp/setens

Berkheya insignis

Brachiaria serrata

Elionurus muticus

Gnidia caffra

Dicoma gerrardii

Jamesbrittenia macrantha

Loudetia simp/ex

Rhynchos;a komatiensis

Rhynchosia totta

Thesium multiramulosum

1

2

47 103

3 4 5 6

91 100 57

17

I I

I

I II

II

II

II

II

II

I

I

I

I

I

I

I

IV

III IV

I III II

IV II

I

II

IV

'"

II

'"

II

II I

I

I

I

I

I

I

I-::-

I

I

I

I f-ro-

III

III

"I

I

I

I

I

I

-

MAJOR GROUP

NUMBER OF RELEVES

SPECIES GROUP P

.. Commelina

africBna

Themeda triandra

Aloe castanea

Pel/aea calome/anos

Indigofer. hi/aris

Raphionacme ga/pinii

Andropogon schirensis

Me/han/a prostrate

M"linis nelViglumis

Tephrosia purpures

1 2 3 4 5 6

47 103 91

100 57

17

I

I

I

I IV III

IV II

III

I

III

III

II

II

II

II

II

V

II

II

SPECIES GROUP

Q

Aristida congest.

Aristida adscensionis

Panicum maximum

Asparagus suaveolens

V

III

I

II

II

II

II

Cynodon decty/on

Kedrostis foefidissima

Po/lichia campestrls

Polyps/a hottentotta

Rue/lis petu/a

Set/der. cspensis

Vernonia fastigiata

II

II

I

II

I

II II

I II I

"

"

"

"

" " "

SPECIES GROUP R

Diagnostic for the Fulrena pubescen~Schoenoplectus

corymbosus

Fuirena pubescens

Andropogon eucomus

I

I

Artemisia afra

Schoenop/ectus corymbosus

Chironia purpurascens

V

IV

IV

IV

III

- l

N

Table 1 continued.

MAJOR GROUP

NUMBER OF RELEVES

SPECIES GROUP R (cont.)

Clffforfia nitidula

Conyza scabrida

Fimbristylis ferruginea

/mperata cylindrica

Miscanthus junceus

Phragmites australis

Plantago lanceoJafa

Puliceri. scabra

Schoenus nigricans

Typha capensis

Verbena brasiliensis

Berula erects

Dfffrichia graveolens

Eucomis autumns/is

var.

clavata

Gomphostigma virgatum

Hypoxis argentea

Ischaemum 'asciculatum

Ky/linga alba

Mariscus congestus

Verbena bonariensis

1

47

2 3

4 5 6

103 91

100 57 17

I

I

III

III

III

III

II

III

III

III

III

III

III

~

MAJOR GROUP

NUMBER OF RELEVES

I

SPECIES GROUP S

~

Senecio microglossus

Cymbopoyon validuB

Hyparrhenia filipendula

Anstida bipartita

Lippia javanics

Lippia rehmannii

Chlorophytum 'asc/cu/atum

Hyparrhenia hirla

Pearsonia obovata

Scabiesa columbaris

Senecio inomatus

Senecio /ygades

Sporobolus centrifugus

SPECIES GROUP T

Heteropogon contortus

Rhus /eptodictya

Eragrosfis cUfVula

Acacia karroo

Eragrostis capensis

Ipomoea obscura

1 2

3 4 5

6

47 103 91

100 57 17

I

I

I

I I

I

I

I

I

I

I

I

II

II

II

II

II

II

III

V

IV

III

III

III

II

II

II

II

II

II

II

II

II

IV

II

II

II

II

II

II

II

II

II

II

I III

II

II

II

II

II

II II

III

III II

III

II III II

II

II

II

II

II

II

III

II

II

Table

2

End~mic, near-endemic and Red Data List taxa recorded for each of the major vegetation types.

-...l w

Species

Acacia karroo [form) (P4)

Acacia sp. nov. (H p.c.)

Adenia wilmsii

Albuca sp. nov. (5856)

Aloe burgersforlensis

Aloe castanea

Aloe pretoriensis

Aloe reitzii var. reitzii

Aloe sp. nov. (5 1419)

Aneilema longirrhizum

Argyro/obium wilmsii

Asclepias sp. nov. (K110)

Asparagus clareae

Asparagus intricatus [form) (W&51501)

Asparagus sekukuniensis

Bauhinia tomentosa [form) (5444)

Berkheya densifolia

Berkheya insignis [form) (5257)

Fam

GIG II GT RI R II Rill R IV RT

FABA e e e

FABA

PASS

LILI

LILI

ASPH n

ASPH

ASPH

LILI

COMM

-

-

-

-

-

-

-

-

-

~: e

-

- Lin

In

-

In

-

-

-

ttl:

ASCL e

ASPA

-

-

ASPA

ASPA

FABA

ASTE

ASTE e

CAPP

CAPP

-

-

e e

-

e e

e e

e

-

-

Ke

-

-

Ke Ke

-

Ke Ke

-

Ke ttl: n n e

-

-

-

-

-

-

-

-

-

-

e

-

e e

-

-

-

-

-

-

-

-

ttl: e e

:ct ttl: e

n n n n

n n n n n n n

n

-

-

-

-

FABA n n

-

-

-

n

n

-

-

e

-

-

-

e e

-

-

-

-

-

-

-

n

-

-

-

-

-

-

-

-

-

Kn n

-

-

-

-

-

-

e e

-

-

-

-

-

-

-

-

-

-

n

-

-

-

-

01 011 aT CI CII CT

-

-

-

-

e

-

-

-

-

-

-

-

e

e

-

e

-

-

-

-

e e

:ct

e e n

-

-

-

-

K n Kn Kn e

-

Boscia albilrunca subsp. macrophylla

Boscia foetida subsp. minima

Brachylaena ilicifolia [form) (5613)

Callilepis Ieptophylla

Catha transvaalensis

Chlorophytum cyperaceum

Combretum petrophilum

Cyphia Iransvaalensis

Cyphostemma sp. nov. A (W13389)

Cyphostemma sp. nov. B (51383)

Cyphostemma sp. nov. C (04142)

Dicliptera fruticosa

Disa rhodantha

ASTE

ASTE

CELA

LILI

C/erodenclrum suffruticosum [form) (51565) VERB

COMB

LOBE

VITA

VITA

VITA

ACAN -

-

e

-

~

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

~

-

-

e e e e e e

N

-

-

-

-

-

-

-

-

-

e -

e e e e e e e e

-

-

e e

e e

-

-

e

e

ORCH - [ [ K

-

-

-

-

- rn1 -

-

-

-

-

n

-

e

-

-

-

-

-

-

-

n n n

-

-

-

-

-

-

-

-

-

-

e e e

-

e e

e

-

-

-

-

-

-

-

-

-

-

-

-

Rn

-

-

-

-

-

-

Rn

-

-

PJRn

-

n

-

-

-

Rn n

-

-

-

AI AT Status

-

-

-

-

-

-

Kn

e

e n n n

-

n e e

-

-

e e

In e

-

-

-

-

-

-

n

e

-

Kn

-

e

Ke

n

-

n

e e

n

7

R

-

-

-

-

-

-

e

-

N

e

n

e

-

n

e

-

-

Rn e

-

e

n

-

K

Table 2

continued

Species

Dombeya autumnalis

Dyschoriste perrotteti

ElephantorriJiza praetermissa

Euclea crispa

[form]

(W&513205)

Euclea Iinearis

[form] (5937)

Euclea

sp. nov.

(W&51686)

Eucomis autumnalis subsp. clavata

Eucomis montana

Euphorbia barnardii

Euphorbia enormis

Euphorbia Iydenburgensis

Euphorbia sekhukhuniensis

Euphorbia

sp. nov.

(W13194)

Gnidia caffra

[form]

(W&512975)

Gossypium herbaceum

Grewia vernicosa

Gymnosporia

sp. nov. A

(W&513351)

Gymnosporia

sp. nov. B (5458)

Helichrysum albilanatum

Helichrysum uninervium

Hemizygia

sp. nov. (5615)

Hermannia antonii

Heurnia insigniflora

Heurnia stapelioides

Hibiscus barnardii

Indigofera Iydenburgensis

Ipomoea bathycolpos var. sinuatodentata

Jamesbrittenia silenoides

Jamesbrittenia macrantha

Jamesbrittenia

sp. nov.

(W13026)

Jasminum quinatum

Jatropha latifolia

var.

latifolia

Kleinia longiflora

[form]

(W&513239)

Fam

STER

ACAN

FABA

EBEN

EBEN

EBEN

LILI

LILI

EUPH

EUPH

EUPH

EUPH

EUPH

THYM

MAlV

Till

CELA

CELA

ASTE

ASTE

LAMI

STER

ASCl

ASCl

MAlV

FABA

CONV

SCHR

SCHR

SCHR

OLEA

EUPH

ASTE

GI Gil

-

-

-

-

n n

I -

-

Ke

GT

RI R II Rill R IV

-

-

-

e n

-

Ke

-

Ke Ke

-

e

e

-

n

-

-

-

-

-

-

-

-

N N N

R

-

-

-

-

-

-

-

e

-

e

-

-

-

-

R

-

R

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

n

~:

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

n

-

n

n

n

t:b:

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

til-

n

-

-

e n

n

n

I e

-

-

-

e

-

e

-

-

-

-

-

-

-

-

-

-

-

-

-

-

:c:t

e

N -

N

K e -

n

-

-

-

-

Ke

-

-

n

n

-

-

-

-

-

-

-

-

-

-

-

-

n e

-

-

-

-

-

e

-

-

RT

01 011

OT

CI CII

CT

AI

AT

Status n

Ke

Ke K e

Ke

Ke Ke

Ke

e e e e

n

-

-

-

-

-

-

-

-

e n

-

-

-

n n

-

Re

-

Re

-

Re

Re

-

Re

n

n n n n -

n

e e e

-

-

-

-

e n e

-

e e e e n

-

-

-

-

n n

e e n

-

-

-

-

-

-

-

-

e e

-

-

-

-

-

-

-

K_e[t

Ke

e

-

-

-

-

-

-

n n n n n

-

-

-

-

n n n n n

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

n n

Ke e n

e e e

-

-

-

e

R

-

lEe Ee

-

....!!...

n

n

Re

-

-

-

-

-

-

-

-

-

e e e

e e

-

e e e e

-

-

-

-

-

-

-

-

N

-

N N N

N

n n n n n n n n n

-

-

-

-

-

-

-

-

-

n e e

-

-

-

n

e

-

-

-

-

n

-

-

-

-

-

-

-

-

-

-

n

-

e

-

-

-

-

~

-

-

-

~

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

N

R

Ee n n

Re e n

N

e

n

-

-

-

-

Ke n

n

-

-

e e e

e

-..J

V>

.

-

Species

Kleinia stapeliiformis

Ledebouria do/omiticola

Leucas capensis

[form]

(W&513007)

Lotononis wilmsii

Melhania randii

[form] (546)

Mosdenia leptostachys

Nuxia gracilis orthosiphon fruticosus

Orthosiphon tubiformis

Ozoroa albicans

Pachycarpus transvaalensis

Pachypodium saundersii

Pavetta zeyheri

[form] (522)

Pegolettia lanceolata

Pegolettia senegalensis

Petalidium oblongifolium

Phyllanthus

sp. nov.

(5470)

PJectranthus venterii

PJectranthus xerophilus

PJinthus rehmannii

Polygala

sp. nov.

(W&513311)

Premna mooiensis

[form]

(W&513004)

Protea caffra

[form] (51382)

Rhoicissus sekhukhuniensis

Rhoicissus

sp. nov. (548)

Rhus batophylla

Rhus engleri

Rhus keetii

Rhus rogersii

Rhus sekhukhuniensis

Rhus tumulicola

var.

meeuseana

f.

pumila

Rhus wilmsii

Rhynchosia nitens

Fam

ASTE

HYAC

LAMI

FABA

STER

POAC

LOGA

LAM I

LAMI

ANAC

ASCL

APOC

RUBI

ASTE

ASTE

ANAC

EUPH

LAMI

LAMI

AIZO

POLY

VERB

PROT

VITA

VITA

ANAC

ANAC

ANAC

ANAC

ANAC

ANAC

ANAC

FABA

GIG II GT RI R II Rill R IV

~:

-

-

-

-

-

-

-

-

Kn Kn

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

n n

-

-

Ke

-

-

-

-

-

-

-

-m-

-

-

-

N e N e

-

-

-

-

-

-

-

-

-

-

-

-

-

e

-

-

-

-

-

-

-mtzJn

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

e

-

-

-

n

-

-

e

-

-

-

e

-

-

-

-

~: e

-

n

N

e

e

-

e

-

n

-

N

-

N N

-

-

n

n

-

-

-

K n -

Kn e

-

-

-

-

-

-

-

-

-

e

-

-

-

-

-

Re

-

-

-

-

-

-

n n

-

Kn Kn

-

-

-

-

-

-

ttl-

K

-

-

-

-

-

-

-

-

-

-

-

-

RT 01 0 II OT CI CII CT AI AT Status

n n n

:r::E:

N

N

~ n

n

Vn

Vn

e

e e

-

e e e

-

-

-

e e

-

e

e

n

-

-

-

-

Ne N e N e Ne

-

-

-

-

-

e e

-

-

-

n

-

-

-

-

n

-

~~

~:

-

n e

-

-

-

-

-

n n n n n n

n n

n

-

-

-

-

-

-

r:-;-

-

-

-

-

n

e e e e e e e e e n

-

-

-

-

-

-

-

-

-

n

-

-

Kn Kn e e e e

~:

-

-

-

e e e e e e

-

-

e e

e

-

Kn

-

e c t J

-

-

K e n

-

-

-

-

-

-

-

-

-

-

-

-

-

-

Ke

-

K

Kn e n

Kn

-

-

n

N

7

-

-

-

-

N

n

~ e

Ne e

-

e e e

-

e e

-

-

Re Re Re

-

Re

-

-

Re

Re Re Re Re n

-

n n n n n n n n

n n n

-

-

n

N

Kn

-

-

-

-

Re

-

-

-

-

-

n

-

-

-

-

Kn Kn

-

-

-

-

-

-

-

-

-

-

-

-

-

Kn

-

-

-

N

K

--l

'"

Table 2 continued

Species

Rhynchosia spectabilis

Schizoglossum

sp. nov. (S628)

Scilla natalensis

Solanum incanum

[form] (W&S13013)

Stapelia gigantea

Stylochaeton

sp. nov. A (S1845)

Stylochaeton

sp. nov. B (S672)

Thesium multiramulosum

Thesium gracilentum

Tragia

sp. nov. (S1573)

Triaspis glaucophyl/a

Tristachya biseriata

Tulbaghia coddii

Tulbaghia

sp. nov. (S1304)

Vitex obovata

subsp. wilmsii

Xerophyta retinervis

[fomn] (W13208)

Zantedeschia jucunda

Zantedeschia pentlandii

SCPE Endemics (e)

SCPE Near-endemics (n)

Total Floristic Elements

Endangered (E)

Vulnerable

M

Rare (R)

I ndetemninate (I)

I nsufficiently Known (K)

Threatened in other re!!ions of s. Afr. (N)

Total Red Data List Taxa

TOTAL

~estricted to the vegetation type

Fam GIG II GT

FABA

ASCL

LILI

SOLA

ASCL

ARAC

ARAC

SANT

~: e

-

N -

-

N

-

-

-

-

-

-

SANT K -

EUPH

MALP n -

POAC

LILI

K

n

~-

K

-

LILI

VERB n -

VELL

ARAC

ARAC

00orr: n e n

e -

~e

Ie

R n Rn

RI R II Rill R IV

n n n

-

I -

-

-

N N

-

e

-

e

-

-

-

-

-

~:

-

-

-

-

-

-

-

-

-

-

-

-

-

n n

-

-

-

-

-

- I K n K n l -

-

-

-

n n n n

e e

-

-

-

-

R n R n

-

-

18 2 19 12 15 2 4

RT 01 0 II OT C I C II CT AI AT Status n e

-

-

-

N N e e

-

-

-

-

-

-

-

N

-

-

-

-

-

-

-

-

e

-

e e e tID-

-

-

-

-

N

-

-

-

-

-

-

-

-

e

e e

e

-

e

-

-

-

-

-

-

-

e

-

-

-

n

N e

N n

-

-

-

-

-

-

-

-

K

-

e e

e e

n n n n n

n

-

Kn

-

-

-

-

-

-

-

-

-

-

-

-

-

-

e

n

-

-

K

Kn

-

-

-

-

-

n n n n n n n e e e e e

e

-

-

-

-

e

n e

-

-

-

-

-

-

-

-

-

Ie

Rn

-

-

-

-

-

-

Rn

23 23 24 31

19

18

28

4

4

52

24 12 15 20 15 16

22

7 7 52

17

1 18 8 20 11 3

20

35

13

7

47 35 39 51 34 34 50 11 11 104

35 3 37

0 0

0

0

2

0

1 0

7 2

5

1

15 3

44 5

0

0

2

1

9

5

17

0

0

0

0

0 0 0 0

1 3 1 0

0 1 1 0

3 3

3

0

1 3 2 0

5 10 7 0

0 0 1 1 0 0 0 0 0

0 0 0 0 0 0

0

1 1

4 3 2 4 1 2 3 1 1

1

1

8

1 0 0 0

0 0 0 0

0 2

5 4 4 6 4 2 4 0 0 15

3 2 1 2 2 1 3 2 2 10

13 9 8 13

7

5

10

4 4 37

50 36

39 52 37 35 54 14 14 121

..!!. 1 4 5 5 7 14 5 5 51

15 2

47

19

20 38 15 7

3 3

0

2

D

=

G Dednam; P

=

PP Swartz; S

=

SJ Siebert; W

=

AE van Wyk sp. nov.

= possibly an undescribed species

Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF

advertisement