Ferns and flowering plants of Klaserie Private Nature Reserve , eastern Transvaal : an annotated checklist

An annotated checklist of the plant taxa of the Klaserie Private Nature Reserve, eastern Transvaal Lowveld, is presented. Of the 618 infrageneric taxa recorded, six are pteridophytes and the remainder angiosperms. Of these, 161 are monocotyledons and 451 dicotyledons. Five of the latter are currently listed in the Red Data List of the Transvaal, two of which are first records for the Transvaal Lowveld. The vegetation of the reserve shows strong affinities with the Savanna Biome, and to a lesser degree, with the Grassland Biome.


INTRODUCTION
No comprehensive plant list o f Klaserie Private Nature Reserve (KPNR) has previously been produced.Witkowski (1983) recorded a total o f 124 taxa, whereas this list comprises a total of 618 taxa.Additions to this list will no doubt be made in the future.Apart from free-rang ing exotic taxa, no other exotic (or indigenous) taxa planted in gardens were included.No algae, mosses, or fungi have been collected or recorded by the author nor, as far as is known, by any other collectors.

STUDY AREA
The Klaserie Private Nature Reserve (KPNR) is situ ated between the town o f Hoedspaiit and the Kruger Na tional Park (K N P), with w hich it shares a common boundary of some seven kilometres.To the east lie the Timbavati, Umbabat, and Ntsiri Private Nature Reserves (Figure 1).
The area was proclaimed as a nature reserve in 1972 and is jointly owned by 107 members.Prior to its proc lamation, the area was partly utilized for cattle fanning.Diseases and predators, however, led to the abandonment of this form of land use.
With a surface area of 62 818 ha, the KPNR is the largest privately owned nature reserve in the Transvaal Lowveld, and probably the largest in the country.Together with the other two large private nature reserves, Timbavati to the east and Sabi-Sand in the south, as well as a number of smaller private nature reserves in the area, it comprises a portion of an important conservation, tourism, and hunt ing area of over 180 (XX) ha.
In terms of a recent Contractual Park Agreement be tween the National Parks Board and a number of privately owned nature reserves, including the KPNR, bordering the Kruger National Park (KNP), common fences between these reserves and the KNP have been removed.
All the more common larger mammals, including the full spectrum of carnivores occurring in the KNP are also found in the KPNR.
Plant specimens were collected or recorded routinely during the course of other work undertaken in KPNR.This formed part of an overall objective of compiling a herbarium collection of the central Transvaal Lowveld re gion outside the KNP Geologically, KPNR consists primarily of granitoid rocks of the Swazian Period, i.e.Makhutswi Gneiss.The gneiss is intruded by a few relatively small, scattered out crops of Harmony Granite and the Phalaborwa complex.Other formations which occur in the area include Milky Quartz Pegmatite and a very small outcrop of Dark Green ish Black Clinopyroxene Hornblende of the Rubbervale Formation (Geological Survey 1986).
The geomorphology of KPNR can generally be de scribed as gently rolling or undulating over most of its area, the southern part being almost Hat.The northern third is somewhat more rugged, however, with many rel atively short drainage lines.Boulder or rock outcrops are common here, in places forming low koppies or ridges.Soils of this region generally comprise shallow sandy loam or gravel, and loose surface stones and rocks are common.Elsewhere, soils are generally deeper and pri marily sandy to loamy or loamy in texture.Along lowerlying areas, clay-loam or clay predominates, particularly along watercourses.Alluvial sand is frequently encoun tered along the banks of the rivers and larger seasonal streams.
D rainage is provided prim arily by the O lifants, Klaserie, and Ntsiri Rivers.The Olifants and Klaserie are normally perennial, though in recent times, particularly during periods of drought, the Klaserie ceases to flow, and the Olifants decreases to a small stream.
Mean annual rainfall recorded in the reserve over the past 16 years totals 424.7 mm.During this period, a mean annual rainfall of 452.8 mm has been recorded at Ncheni Gate, 408.5 mm at Xanatseni Gate, and 432.0 mm at Xiplapaleni Gate.A mean of 407.8 mm over a 33-year FIGURE 1.-The locality of the Klaserie Private Nature Re serve.Broken lines denote quarter-degree squares.Inset shows the reserve's locality within the Transvaal.Biomes after Rutherford & Westfall (1986), figure taken from Huntley (1989).
period has been recorded at the reserve's headquarters (Warden's records, unpublished).
The reserve's mean annual rainfall is well below 530 mm, the approximate upper limit for the arid subdivision of the Savanna Biome, and is close to 400 mm, the lower limit given for the dry subdivision of the Grassland Biome (Rutherford & Westfall 1986).
Altitude ranges from 303 m in the northeast at the con fluence of the Klaserie and Olifants Rivers, to 535 m in the southwestern comer-a difference of 232 m over a distance of 33 km.
The vegetation of KPNR falls within the Savanna Biome (Gibbs Russell 1987;Rutherford & Westfall 1986).According to Acocks (1988), the only Veld Type of the reserve is Arid Lowveld of the Tropical Bush and Savanna Types (Bushveld).In the northeastern and eastern sectors of the reserve however, mopani veld (Colophospermum mopane) is present and is dominant over large areas, de creasing in extent southwards, where it occurs in the form of small, scattered pockets.The vegetation of KPNR has been mapped in greater detail at the reconnaissance level, and 15 m ajor associations w ere identified (Zambatis 1983).

METHODS
The great majority o f taxa given in the list which fol lows were collected or recorded over the period 1973 to 1987 by the author whilst in the service o f the Transvaal Directorate o f Nature and Environmental Conservation (TDNEC).Apart from duplicate specimens donated to the National Botanical Institute's National Herbarium in Pre toria (which undertook the identification o f all the speci mens collected), the collection is housed in the herba rium o f the Hans Hoheisen Wildlife Research Station (HHWRS), bordering the Kruger National Park at Orpen Gate.
Taxa recorded but not collected by the author were listed after being identified by reference to herbarium ma terial collected elsewhere in the region.Where no doubt existed as to their identity, these were recorded without any reference to herbarium material.
N om enclature, taxonom ic sequence of arrangement and spelling used throughout this list follows that of the National Botanical Institute's PRECIS system as detailed in Gibbs Russell et al. (1985Russell et al. ( , 1987Russell et al. ( , 1988;;De Wet et al. 1989, 1990, 1991;Arnold & De Wet 1993).Taxonomic numbering has however been omitted in order to enhance clarity and brevity.
Synonyms are given only in those cases where taxo nomic revision and subsequent name changes occurred after specimens were originally named.
With few exceptions, the taxa have been annotated with summarized field notes made at the time of collec tion.A standard sequence of annotation has been followed whereby the first category represents the life form, fol lowed by the height range, vegetation type, habitat, and soil texture.These categories are separated by a comma.Where a category was not recorded, this is indicated by a dash.
The vegetation and habitat types given for each taxon refer only to the specimens collected or recorded.This does not imply that the taxon in general is restricted to this vegetation or habitat type.Many of the taxa occur in a variety of vegetation and habitat types.
Life form classes are according to the Raunkiaer sys tem as used by Muller-Dombois & Ellenberg (1974) and Rutherford & Westfall (1986).Height classes of phanerophytes are those of Muller-Dombois & Ellenberg (1974).The following classes are used: Phanerophytes (P): perennial plants, usually woody, mean height of the renewal buds > 0.7 m above ground.Mesophanerophytes (Me): mean height of the renewal buds 5 -5 0 m.Microphanerophytes (Mi): mean height of the renewal buds 2-5 m.Nanophanerophytes (N): mean height o f the renewal buds < 2 m.Chamaephytes (Ch): perennial plants, generally woody or partly woody, mean height o f the renewal buds < 0.7 m.Hemicryptophytes (H): perennial plants, generally herbaceous, renewal buds at, or usually, close to ground level and seldom over 0.1 m.Geophytes (G): perennial plants, usually herbaceous, renewal buds below ground level.T herophytes (T): ephemeral plants (annuals).Lianas (L): plants that grow by supporting themselves on others.Epiphytes (E): plants that germinate and root on other plants, including dead standing plants.Parasites (Pa): green plants growing at tached to other living autotrophic plants.
The other annotations are explained below: Height: given in metres.In most cases, a height range is given, though where only one specimen was collected or recorded, only one height is given.Vegetation type: the dominant (i.e.most abundant) woody species, such as Combretum, Acacia, Colophospennum, etc. 'Mixed wood land' is used in cases where dominance by one or more species is not clearly evident and is thus a mixture of various species.The term 'woodland' is used in a broad context and does not necessarily imply a dominance by trees.A common feature in fact is the presence of both trees and shrubs, in varying degrees of dominance, though in no case where 'woodland' is used, is the tree stratum absent.'Riverine' refers to the woodland and shrubveld along the two perennial rivers, the Olifants and Klaserie, whereas 'seasonal stream' refers to the vegetation occur ring along the banks of the numerous watercourses of a seasonal nature, varying from substantial rivers such as the Ntsiri, to the many minor streams, gullies and other tributaries of the perennial rivers.In both these forms of riverine vegetation, physiognomic variations exist and range from poorly developed, to closed tall woodland, with or without a dense (or more open) thicket under storey.Although many taxa are common to both of these riverine forms, sufficient physiognomic and floristic dif ferences in these (and in the water regime) exist to war rant a separation between them.'Koppie' is applied to a range of vegetation types associated with rocky ridges, koppies, or isolated boulder or rock outcrops of varying extent and height above the surrounding terrain.'Open parkland' refers to the vegetation structure of the seepline complex where trees are scattered or absent, with a scat tered shrub stratum, whereas 'grassland' refers only to sea sonally or perennially waterlogged marshy areas (vleis), the presence of woody plants being a rare exception.Hab itat: the term 'dryland' refers to the broadly homogeneous area occurring between one drainage line and another, but excludes distinctly different or specialized habitats which are floristically and physiognomically atypical of the sur rounding area and which generally occupy the same to pographic position.Examples of these exceptions are koppies and rock outcrops, seasonal seepage areas, vleis, and riverine habitats.An overriding feature of dryland is the presence of a woody component in the form of one or two strata, namely tree and shrub layers, each occurring in varying frequency or dominance, ranging from sparse to dense.'Dryland' is thus synonymous with 'veld', in a broad context, but within the confines of the Savanna Biome of Rutherford &Westfall (1986) andGibbs Russell (1987).
A 'seepline complex' represents a specialized habitat for a variety of plant types, a number of which appear to be restricted to this type of habitat.These areas are com monly located on slopes on a catenary sequence of soil types.Sandy soils occur on the upper side of the seepage area, grading into sandy loam and finally into clayey loam or clay at the bottom of the slope.During the rainy season, and particularly after several days of heavy rain, water seeps to the surface at the sand/clay interface, creating damp or even waterlogged conditions for prolonged peri ods.A gradient in the distribution of plant types along these areas is also evident, with Terminalia sericea char acteristically demarcating the upper edge.These areas are vulnerable to erosion when heavily utilized by herbivores due to the concentration of salts which lead to defloccula tion and collapse of the clay structure.When in this state, they are known as sodic areas (Scholes 1985).In this ar ticle, no distinction has been made between the two, as certain plant taxa occur on both types of habitats, hence the term 'complex'.
A 'rock outcrop' refers to the habitat structure of the koppie vegetation type and can comprise large and exten sive boulders, or isolated clumps of rocks.In practically all rock outcrop habitats, soils are shallow to very shallow, and sandy or gravelly, with smaller, loose stones being common.

RESULTS AND DISCUSSION
A total of 618 infrageneric, or 'lower', taxa have been recorded in the KPNR.This does not include specimens identified to the genus level only, except those which are the only members recorded in a particular family or genus.Taxa of the same species but with different subspecies or varieties were regarded as different taxa.These are summarised in Table 1 according to the number of fam ilies, genera, and taxa in the pteridophytes, monocotyle dons, and dicotyledons.
Twenty-seven (26.2%) of the total number of families contribute 1% or more of the total number of taxa (Table 2).Genera with six or more taxa (> 1 % of the total number of taxa) are listed in Table 3.
Free-ranging exotics are represented by 39 taxa.
These represent 2% of the total number of 251 taxa currently listed in the Red Data List for the Transvaal.Two of these, N. lotus and C. mafekingensis, are first re cords for the Low veld (E.van Hoepen pers.comm.).Nei ther oi these taxa have yet been collected in the KNP, though N. lotus has apparently been recorded by S.P. Fourie of the TDNEC.In the case of A. vandermenvei, the only specimen collected was from a small group which appears to have been planted at Xanatseni Gate.A search for wild-growing specimens in the vicinity of this gate proved fruitless.It is possible that these specimens were introduced into the reserve from beyond its boundaries.This plant thus remains to be confirmed in its natural hab itat within the reserve.
Despite its relatively arid climate, and the fact that the surface area of KPNR is a mere 0.1% of the total surface area of the Savanna Biome of 632 034 km. the reserve's  1987).The riverine habitats along the two perennial rivers of the reserve partly account for this relatively high num ber of taxa.If the frequency and duration of desiccation of these rivers, however, increases in the future (as it is likely to do, given the ever-increasing human demands on all water resources), a decline of these plant communities, or even the disappearance o f certain taxa is likely to occur.
The quarter-degree squares o f the KPNR were not in cluded in the PRECIS search undertaken by Gibbs Russell (1987) in her analysis o f the southern African biomes, and in the determination of the core areas of these biomes.In this search, she recorded 21 families, comprising 1% or more of the taxa, which together account for 55-60% of the total number of taxa recorded in the Savanna Biome.These families, all of which have been recorded in KPNR as well, account for 459 (74.2%) o f the reserve's total number o f taxa.These relationships consequently reflect a strong affinity of the flora of KPNR with that of the Savanna Biome in general.
A total of 22 families and 36 genera with 10 taxa or more, have their centres of diversity in the Savanna and Grassland Biomes (Gibbs Russell 1987).All except one of these (Aspleniaceae) and six genera, have also been recorded in the KPNR.indicating a strong relationship with the Grassland Biome as well.It is significant to note though, that the Orchidaceae, represented by a total of 53 taxa common to the Savanna and Grassland Biomes, and which, together with the Lam iaceae, distinguishes the Grassland Biome (Gibbs Russell 1987), is very poorly represented in KPNR, with only two taxa being recorded.This poor representation of the Orchidaceae can most probably be ascribed to the relatively arid climate of KPNR.
In the Umfolozi Game Reserve, on the other hand, this family is absent, in spite of a mean annual rainfall of 625 mm.Downing & Gibbs Russell (1981) suggest that wild herbivores could have severely depleted the rhizomes of any ground orchids present, while extensive and intensive spraying of insecticides during the tse-tse fly eradication campaign could have inhibited sexual reproduction by eliminating insect pollinators of both ground and epiphytic orchids.A similar situation is evident in the Manyeleti Game Reserve, bordering the KNP on the western side of the central region of the Park.Here, Bredenkamp (1982) only recorded one taxon of this family, even though this reserve falls within the mesic subdivision o f the Savanna Biome (R utherford & Westfall 1986).No insecticide spraying has been undertaken in KPNR.
The generally arid nature o f a m ajor part o f the Lowveld consequently appears to be the reason for the relatively poor representation o f this family in this part of the Savanna Biome, which therefore weakens the link be tween this biome and the Grassland Biome somewhat, at least in this region.
The aridity of KPNR is most probably also the reason for the absence of the Aspleniaceae.This large family ap pears to be confined to the higher rainfall regions of the country, as shown on the distribution maps of Jacobsen (1983) and Burrows (1990).This is further reflected by the fact that the only two members of this family recorded in the KNP are restricted to the Pretoriuskop area of the Park, which receives the highest mean annual rainfall for the Park of some 722 mm.Table 4 summarises the life forms of the KPNR flora.This shows that woody plants-phanerophytes and chamaephytes, together comprise 31.7% of all taxa, whereas herbaceous plants, the hemicryptophytes and therophytes, comprise 51.7%.The high frequency of woody and her baceous taxa is further evidence that the vegetation of the reserve comprises a woody layer above an herbaceous layer.

CONCLUSION
The primary purpose of this article is to provide a sys tematic list of the flora of KPNR, which can be used in a variety of more detailed studies.Nevertheless, this rather brief analysis of the reserve's flora shows that very strong affinities exist with the Savanna Biome, and to a lesser degree, with the Grassland Biome as well.The vegetation of the Klaserie Private Nature Reserve can therefore be regarded as comprising part of the core area of the Sa vanna Biome.

ACKNOWLEDGEMENTS
The many landowners of KPNR are thanked for al lowing me to collect plant specimens on their properties, as is the Warden, Mr E.P Leibnitz, for his interest and support.
The identification of the large number of specimens collected over the years was undertaken by the National Botanical Institute, Pretoria.Without the assistance of the many persons involved, it would not have been possible to compile this list.In particular, the assistance of Mrs E. van Hoepen, Ms E. Retief, and Mrs B.C. de Wet is grate fully acknowledged.
The Directorate of Nature and Environmental Conser vation (Transvaal) is thanked for providing the support which made this work possible.

SYSTEMATIC CHECKLIST
After the name of the author(s) of the species name, the sequence of annotation is as follows: collection num ber: without initials-author; with initials-RPE = R.P Ellis, JdK = J.PS. de Kock.P.J.M. = P.J.M. Muller.

TABLE 2 .
-Families with 1% or more of the total number of infrageneric taxa