Exploding the myth

Grassland in the Drakensberg foothills with a lone Cabbage Tree (Cussonia paniculata). Photo: Eugene Moll.

Fire and frost keep trees from establishing in the grasslands of the Highveld is an oft-perpetuated myth. Back in the late 1950s, an exceptional South African geomorphologist, Lester King, published his book The South African Scenery in which he questioned many existing geological paradigms, and Ken Tinley put forward his view of why the Highveld was treeless. Tinley’s view was certainly very different from what I had been taught, and what is still taught today – that frost and fire keep the Highveld grasslands treeless. What rubbish!
As I remember it, the reason for King’s book was that he had submitted a paper on the age of South African mountains and the peer-reviewed journals (in the Northern Hemisphere) all refused to publish his work. At that time geologists believed that there were no mountains on Earth older than 10 million years! Yet King was telling the world that the Cape Fold Mountains were older than 70 million years old and that we had some land-surfaces that hark back to the break-up of Gondwana (the Highveld). Today his pioneering work has been vindicated. Thankfully today’s scientists have become much more open-minded since those oppressive and restrictive days of the mid-1900s. Recently a University of Cape Town group researched the reasons for no trees in the Highveld grasslands (see the article ‘Why do grasslands have no trees’ by Julia Wakeling, William Bond and Michael Cramer in the March 2009 issue of Veld & Flora). Their conclusions were inconclusive, but they certainly demonstrated that frost is not a totally limiting factor, nor were fires a conclusive reason for no trees but that the lack of trees could be due to a combination of factors; one of which may be plant available nutrients. Thus the generally held view that grasses dominate the Highveld because it is too cold for trees and that fierce annual fires kill woody growth has been debunked.

Ok then what are my reasons for the lack of trees? Firstly we have to understand that most of the Highveld is a Gondwana land-surface 70 million years old. Soils have been leached of most of their important plant nutrients. The relative quantities of aluminium (one of the most common elements in the Earth’s crust) are very high, and this could be a factor in restricting root exploitation, Tinley concluded. Tinley had
observed that alien trees had no problem growing in the Highveld and were in fact invasive (Australian wattles and gums, and certain species of pine – all of which evolved in extremely toxic aluminium soil environments, so were pre-conditioned to survive in the Highveld).
Tinley also observed that some indigenous trees could grow well, but only in certain places where the sub-soil drainage had been affected by, for example, incipient dongas and roads. Thus he debunked the frost and fire theory way back in the 1960s. Plants get most of their nutrients from the top few centimetres of soil where almost all the feeding action takes place. Grasses are basically shallow rooted and trees (be they seedlings, saplings or mature individuals) compete with grasses for 99% of their nutrients in the organic-rich top soil. However, trees, unlike grasses, send down anchor and water-gathering roots much deeper – into that toxic soil horizon. Thus in relatively undisturbed grasslands the grass sward is such an extremely competitive environment that it is difficult if not impossible for any seedling to survive. Even grass seedlings have to wait for some episodic event to cause a small gap. Another fact is that unlike savannas, where there are all kinds of seeds of woody plants in the soil waiting for an opportunity to germinate and establish, in the Highveld there is no such soil stored seed-bank. So even if the grass sward integrity is breached there are no tree seeds waiting to spring into action. Today not only has the grass sward been impacted but local drainage patterns have changed, thus the whole grass competitive top soil and underlying toxic soil integrity has been impacted paving the way for more tree invasion – including indigenous species. If Tinley is right then one would expect to see changes in the establishment pattern of woody plants in the Highveld grasslands
today. I have been travelling through this area since the 1960s and l see many more indigenous trees in these grasslands. This has not surprised me as in my ecological teaching I have always given students the ‘Tinley view’ in opposition to the ‘classic view, now disproved’. The current rate of invasion is more rapid, and that fits with the Tinley view as most of the Highveld has now been disturbed and because of the planting of indigenous trees along roads and in gardens, the seed source is not as remote as it once was. Thus within a few years native tree species will be a much more common sight in these Highveld grasslands, and the next generation will not see these once vast treeless Highveld grasslands. Just as my generation have been denied the opportunity of seeing the vast herds of grazing antelope that one roamed these vast grasslands.
Eugene Moll, Cape Town

Julia Wakeland replies
It is fantastic that this article has stimulated this and many other responses to one of South Africa’s most obvious yet unexplained ecological patterns. Tinley, Feely, Acocks and others have suggested reasons to explain the lack of trees in grasslands, but none of them are sufficiently universal to explain the biome as a whole. For example, adverse soil conditions such as seasonal water-logging or aluminum toxicity may prevent trees from growing in some areas, but there are also grasslands without seasonal water-logging or high aluminum levels. There is scarce experimental work tackling this question, which is unsurprising given that fieldwork over such large areas is logistically overwhelming (speaking from experience). However, we delight in
experimental work as it provides evidence that may force us to change our minds.
In our experiment we quantified tree growth rates in savanna and grassland. Firstly, trees did grow in grassland soils. Nutrient differences meant that trees grew slightly slower in grassland soils than savanna soils. However, temperature differences between the two biomes had a much larger effect on tree growth. In combination with frequent fires the slow growth of trees in grassland temperatures would prevent trees from establishing as adults. In this experiment we were interested in the sapling to adult life stage. With our research group (the Zululand
Tree Project) we have demonstrated that seedling growth is adversely affected by grass competition in savannas, and are currently doing experimental work to explore the competition of grasses in grasslands.
Our focus was on savanna trees that survive grass competition and frequent fires in savannas. There are many fire-sensitive trees such as White Stinkwood (Celtis africana) that typically grow in forests, and would not be expected to grow in a grassy environment. The real puzzle
is why the tree-grass mixture of savannas becomes grassland.
We feel like we are slowly homing in on the answers to this amazing phenomenon of a savanna tree-line well below the altitude of an alpine tree-line, but are not there yet. If you have any brilliant ideas that we haven’t thought of yet we would love to hear them! If you’d like to read more, a brilliant starting point is Tim O’Connor and George Bredencamp’s ‘Grassland’ chapter in the book Vegetation of Southern Africa, edited by R.M. Cowling, D.M. Richardson and S.M. Pierce.