Weed flora of South Africa 3 : more power shifts in the veld

This paper deals with shifts towards plants with particular life cycles, growth forms, or from particular geographic areas. Exotics probably have their greatest impact in the aquatic habitat where they almost equal in number of species (the ratio is 3:5) and out-perform the indigenous aquatic flora. In the streambank habitat exotic weed species, mostly trees, outnumber indigenous weed species by more than 3 to 1. They are completely replacing indigenous streambank communities in many places. An investigation of their effect on stream flow and water loss is urgently required. The problems of the winter rainfall area are highlighted by the fact that it has 300% more indigenous species and 109% more weed species per unit area than the summer and all year rainfall areas. In the veld as a whole there is a significant and so far unremarked invasion by exotic annuals. The major invasive weed groupings are herbs from Europe and Asia and trees from Australia but South America probably has the greatest potential to provide us with new and dangerous weeds. A comparison of the weed floras of the southern continents could contribute much to an understanding of our own weed flora.


INTRODUCTION
If it is true to say that scientists need to be as international as the material on which they work, then weed scientists should be the wandering Jews (or Arabs) of the science world.Certainly, it is impossible to work on South African veld weeds without becoming acutely aware of what is indigenous and what exotic.This harping on what is one's own and what is not, may seem strange to botanists from Europe where wave after wave of invading peoples and thousands of years of cultivation have mixed and blended, leaving an international flora.But the kingdoms of the south, the floras of Africa, America and Australia have evolved in isolation from one another, and from the great temperate plant reservoir of the northern hemisphere.These floras have unique characteristics and it is only in the last few hundred years that they have been brought together here at the southern tip of Africa.When they meet, it does not take a botanist or a poet to see them as green armies doing battle -that is exactly what they are.
With the battle-lines so clearly drawn, with our own flora one of the richest and most vulnerable in the world (Oliver, 1978) and with a fighting chance to save it, there is some justification for our preoccupation.
This paper deals with various habitats in the veld and with shifts towards plants with particular life cycles, growth forms, or from particular geographic areas.The weeds referred to are all in the National Weed List (Balsinhas et al, 1982).

LIFE CYCLES
The 966 exotic and indigenous weeds of the veld have been classified according to life cycle (Table 1).
The great majority of our weeds of the veld (particularly the indigenous ones) are perennials.The lack of indigenous annual weeds may be a real

HABITAT AND GROWTH FORM
The 1 003 exotic and indigenous species that are weeds in the veld and in water bodies have been classified according to habitat and growth form (Table 2).A number of them occur in several habitats, giving a total of 1 334 species for all habitats.

Aquatic habitats
The scarcity of water is the greatest limiting factor in the development of vast areas of South Africa.It is so valuable and large water-bodies suitable for recreation are so few that almost all plants growing in water are somewhere regarded as weeds.The 25 indigenous water weed species listed constitute almost our total indigenous aquatic vascular plant flora.They are mostly rooted species adapted to running' water or to shallow waters of vleis and pans.
In terms of their percentage contribution to the flora of the habitat exotics probably have their greatest impact in the aquatic habitat, where there are almost as many exotic as indigenous species (the ratio is 3:5).However they do not always compete directly with indigenous species, and few indigenous aquatic species are endangered.This is because most of the exotics flourish best in standing and deep water situations, including man-made dams and canals.

Streambank habitats
The next greatest area of impact of exotics is probably the streambank-habitat, where exotic weed species outnumber their indigenous counterparts by more than 3 to 1 (Table 2).Most of these exotic invaders are woody trees and shrubs -growth forms that have a maximal effect on their neighbours, on lower strata and the habitat as a whole.Indigenous streambank communities are being completely replaced in some parts (Wells et al., 1980) and there is an urgent need to establish what effect the new exotic communities are having on water loss and water flow.
The aquatic and streambank habitats occupy only about 1% of the area of the country but they are key areas where weeds have a high impact.

Terrestrial habitats
The terrestrial habitats that occupy 99% of the country are split as follows: -the winter rainfall area covers about 11%, houses about 30% of our indigenous species, and about 18% of our weed species (both indigenous and exotic) -the summer and all year rainfall area covers about 88% of the country, houses about 60% of our indigenous species and about 69% of our weed species.
The winter rainfall area has about 109% more weed species and 300% more indigenous species of all kinds per unit area than has the summer and all year rainfall area.If one realizes that most of the winter rainfall area has been drastically changed as a result of agriculture and urbanization etc., and that the rich Cape flora is concentrated in relic patches totalling 1,8 million ha (Hall et al., 1980); that over 90% of these species are endemic (Oliver, 1977); and that they include the highest concentration of threatened species in the country, then the flora weed problem of the area is put into frightening perspective.
Terrestrial weeds are very evenly divided between exotic and indigenous species in both the winter and summer rainfall areas.The ratio of different growth forms also varies little with a lower percentage of dwarf shrubs and climbers and a higher percentage of trees and shrubs amongst the exotics.It is the exotic trees, particularly Hakea, Acacia and Pinus species, that have wrecked havoc in the Cape fynbos where a natural tree layer is lacking (Oliver, 1977;Hall et al., 1980).

REGION OF ORIGIN
The region of origin of the 481 naturalized exotic weeds (flora weeds) and of the 16 exotic water weeds is shown in Table 3.A number of them occur in several habitats, giving a total of 685 species for all habitats.
Europe and Asia supply the largest number of flora weeds (40% of the total for all habitats).Of these 73% are hemi-cryptophytes or floating species and the region supplies 47% of all our flora weeds in this growth form category.
South America is the second largest contributor of flora weeds overall.Its species are well represented in all habitat and growth form categories and it contributes by far the most (60%) of the climbing and dwarf-shrub species, as well as the most water weeds.
Africa (and other) supplies mainly hemicryptophytes and water weeds of the summer rainfall area.
Australasia supplies only 11% of our flora weed species but of these a remarkably high proportion (81%) are trees or shrubs.They are particularly important in the winter rainfall area but many of the species are invasive in all three non-aquatic habitats.
Central and North America is the least well represented region supplying mainly hemicrytophytes of the summer rainfall area.
At present it is not easy to know to what extent these patterns reflect the aggressive potentials of the other floras now interacting with our own.We have already suffered from the great influx of herbs from Europe and Asia, and of woody species from Australia.Our impression is that the greatest potential for future weed invasion is South America which has many hardy grasses like the Stipas, shrubs like Larrea and trees like Prosopis that have yet to penetrate to South Africa.
A comparison of the weed floras of the southern continents could contribute much to an understan ding of our own weed flora and the hazards that await us in the form of new introductions.

CONCLUSIONS
Exotics probably have their greatest impact in the aquatic habitat where they almost equal in number of species (the ratio is 3:5) and out-perform the indigenous aquatic flora.In the streambank habitat exotic weed species, mostly trees, outnumber indigenous weed species by more than 3 to 1.They are completely replacing indigenous streambank communities in many places.An investigation of their effect on stream flow and water loss is urgently required.
The problems of the winter rainfall area are highlighted by the fact that it has 300% more indigenous species and 109% more weed species per unit area than the summer and all year rainfall areas.In the veld as a whole there is a significant and so far unremarked invasion by exotic annuals.
The major invasive weed groupings are herbs from Europe and Asia and trees from Australia but South America probably has the greatest potential to provide us with new and dangerous weeds.A comparison of the weed flora of the southern continents could contribute much to an understan ding of our own weed flora.

TABLE 1 .
-Life cycles of indigenous and exotic weeds of the veld

TABLE 2 .
-Growth forms of exotic and indigenous weeds in veld and water * In columns A and B 100% = total species in each habitat; in column C 100% = total species in all habitats.

TABLE 3 .
-Regions of origin of flora weeds of various growth forms and habitats * For total flora weeds = 100% in each growth form category see Column B of Table2.