Numerous researchers working throughout the geographic range of Australian rainforests have advanced the theory that rainforest boundaries are controlled by fire. This is an old idea, possibly first advanced by the Czech botanist Domin (1911). At the heart of the theory is the assumption that rainforest tree species are ‘fire tender’. Francis (1951) wrote that ‘one of the most marked differences between the constituents of rainforests and those of the open Eucalyptus forests is their behaviour towards fire. In most if not all cases the rainforest constituents are killed even by slight contact with or proximity to the fires which periodically sweep through many of the Eucalyptus and open forests of Australia’. Similarly, Webb (1968) argued that the difference in the ability of rainforest and non-rainforest vegetation to tolerate fire ‘is reflected in the remarkably sharp boundaries of fire-sensitive raingreen forests in the tropics and subtropics, which is related to the exclusion of fire, virtually on an all-or-nothing basis’. However, the supporting evidence is limited and often circumstantial. My aim in this chapter is to review the Weld evidence that fire is critical in controlling rainforest boundaries throughout Australia. Evidence from the humid tropics, monsoon tropics, subtropics and temperate regions will be considered in turn.

Humid tropics

Webb (1968) argued that, in the tropics, rainforest and non-rainforest vegetation is in a dynamic balance controlled by fire frequency, soil fertility and topographic settings such as ‘rocky outcrops and gullies, especially in the lee of fire-bearing winds’.

Introduction

An obvious feature of the boundary between rainforests and adjacent nonrainforest formations is a dramatic change in microclimate. Seddon (1984) suggested that microclimate is often unconsciously used to dichotomise rainforest fromnon-rainforest in Australia. He wrote that rainforests are ‘the only forest form that is at all common in Australia that does cast a dense shade’ being ‘so different from the familiar, light-drenched open forest dominated by eucalypts’ (original emphasis) (Figure 7.1). It has often been assumed that the differences in microclimate are of significance in controlling the establishment of seedlings. Herbert (1932) wrote that ‘in the struggle for existence in the crowded rainforest, sun-loving types, such as Eucalyptus, have no chance of becoming established’. Another factor that changes across the rainforest boundary is ground surface temperature. Minimum temperature and frost are also thought to prohibit subtropical and tropical rainforest from occurring on otherwise optimal sites (Webb and Tracey 1981) and past colder climates may account for the occurrence of grasslands within tracts of subtropical rainforest (Webb 1964). The purpose of this chapter is to:

1. (i) describe the difference in light across rainforest boundaries;

2. (ii) evaluate the hypothesis that differences in light regime control regeneration of rainforest and non-rainforest trees;

3. (iii) consider the effect of maximum and minimum temperature on rainforest distribution.

In Chapter 8 I summarised a large body of Weld evidence, collected throughout the geographic range of Australian rainforest, that strongly suggested that fire plays a major role in controlling rainforest boundaries. However, I also showed that the relationship between fire and rainforest distribution is not simple. Rainforest trees are best differentiated from non-rainforest tree species by their inability to survive recurrent fires; most rainforest tree species can regenerate after a single fire. Thus, the frequency of fires is critical in controlling rainforest. In the previous chapter, I showed that differences in soil fertility across rainforest boundaries reflect the cumulative effect of fire history. In an influential paper published in 1968, Jackson emphasised the role of fire frequency in determining soil fertility and vegetation types in western Tasmania. Specifically, he developed the ‘ecological drift’ theory that emphasised the importance of variation in fire frequency in determining the distribution of vegetation types, including rainforest, in western Tasmania. Has Jackson found the key to understanding the distribution of Australian rainforest? In this chapter, I critically review Jackson's theory.

Rainforest, succession and the ecological drift model

Jackson (1968) set out to answer a simple question: why should rainforest be spatially restricted in western Tasmania? With its cool, mid-latitude oceanic climate and an annual rainfall in excess of 2000mm distributed throughout the year, western Tasmania would seem ideal for cool temperate rainforest. However, rainforest is of limited extent and treeless sedgelands dominated by Gymnoschoenus sphaerocephalus (commonly known as button-grass) are widespread (Figure 10.1).

In Chapter 5 we followed the processes of reproduction through to the arrival of the seed on the ground. Here we will look at germination and early survival of the seedling, and ways of producing new trees without resorting to seed.

The Seed

Seeds remind me of spaceships: they contain everything they need to colonise new worlds given favourable conditions and water once they arrive. The outside is covered by the seed coat (the testa), designed to protect the contents (see Figure 5.13). At the centre of the seed is the embryo, consisting of little more than a miniature root (the radicle) and shoot (the plumule). The rest of the seed is taken up with the food supply to keep the embryo alive before it germinates and to sustain early growth before photosynthesis takes over. This food is stored usually in the cotyledons (seed leaves), although some store it outside the cotyledons in the endosperm (which can be thought of as a short-lived half-brother of the embryo; all flowering plants have endosperm but in most it is used up quickly). In ash (Fraxinus excelsior), for example, the cotyledons are small, surrounded by endosperm, but in oak (Quercus spp.) the cotyledons are bloated and fill the seed with no remaining sign of the endosperm. The best example of endosperm is in the coconut (Cocos nucifera): part liquid (the milk) and part solid (the flesh).

A common question is to ask ‘how quickly will my tree grow’ or ‘how big will my tree eventually get?’. In this chapter we will look at these and related questions, and the reasons behind the answers.

Speed of growth

Height

You have probably seen films where the hero in the Orient is strapped over a bed of growing bamboo as a means of torture and eventual death, speared by the hard growing shoots. The reason this works is the extraordinarily fast growth of over half a metre per day. While tropical vines and lianas can grow almost as fast, trees proper can't equal this rate but can nevertheless be impressively quick, especially when young. A number of tropical species can add 8–9 m to their height in a year. A New World relative of the elm, Trema micrantha, has been seen to grow 30 m in 8 years (an average of 3.75 m per year) and a eucalypt (Eucalyptus deglupta) in New Guinea reached 10.6 m in just 15 months. The Guinness Book of Records quotes the air-speed record for a tree as a specimen of Albizzia falcata planted in Malaysia, which grew 10.74 m (35 ft 3 in) in 13 months! As you would expect, there is a lot of variation between species. Trees that invade gaps in tropical forests, and need to grow quickly to win the race to the top, grow faster (an average of 1.5–4.0 m in height per year) than later species that can afford to slowly plod upwards through the shade (0.5–1.2 m per year).

A common view of tree roots is that they plunge deep into the ground producing almost a mirror image of the canopy. Yet in reality a tree looks more like a wine glass with the roots forming a wide but shallow base (Figure 4.1). Most trees fail to root deeply because it is physically difficult and unnecessary. The two main functions of roots are to take up water and minerals, and to hold the tree up. In normal situations, water is most abundant near the soil surface (from rain), and this is also where the bulk of dead matter accumulates and decomposes releasing minerals (nitrogen, potassium, etc.). It should not be surprising, therefore, to find that the majority of tree roots are near the soil surface. The flat ‘root plate’ also serves very well for holding up the tree; deep roots are not needed (see Chapter 9).

Roots have other functions. They store food for later use (see Chapter 3) and they play an important role in determining the size of the tree. Roots normally account for 20–30% of a tree's mass (although it varies from as little as 15% in some rainforest trees up to 50% in arid climates). If the trunk is ignored (40–60% of the total mass), however, the canopy and the roots come out roughly around the same mass. This helps put into perspective the relative value of the roots and the leaves to each other.

The whole point of a woody skeleton is ultimately to get the leaves above competitors to ensure a lion's share of the light. And from this simple goal comes an enormous range of tree shapes, from the unbranched stems of palms and tree ferns to the tall spires of conifers, the broad spreading crown of oaks and the multiple stems of an old yew. What governs the shape of trees? How are trees organised to display what often looks like an impossibly large number of leaves?

Trees of distinctive shape

It is usually possible (but not always!) to identify a conifer from a distance by its conical outline. Within the cone there are usually plates of foliage showing where the branches are produced in whorls around the main central stem, usually one whorl per year (Figure 7.1). This contrasts with the rounded dome of a hardwood where the initially leading shoot of the young tree gives way to a number of strong branches, giving the whole canopy a rounded shape.

Within these two main shapes it is possible (with a little practice) to distinguish different species simply by their shape. This book is not the place to list the distinctive features of common species but one example will illustrate the point. In common lime (Tilia × europaea) the main branches develop in great arching curves, which in time lose the terminal buds.

It's a tough world. Trees face a constant battle in competing for light, water and minerals with surrounding plants. As if that were not enough, they also have to fend off the attention of living things, which view trees as good to eat and places to live. Insects chew away on all parts of a tree and are quite capable of completely defoliating it. Larger leaf-eating animals (which are usually on the ground since a belly full of compost heap is a heavy thing to carry around; leaf eating monkeys are an exception) chew away at the lower parts of the tree, although giraffes can reach up around 5.5 m. Whole armies of animals that can climb and fly will feed on the more nutritious flowers, fruits and the sugar-filled inner bark (see Chapter 3). The grey squirrel, introduced to Britain from N America in the 1880s, is a prime example. This rodent does extensive damage to hardwoods by stripping bark in spring to get at the sweet sap. It seems that dense stands of selfsown hardwoods have little sap and are largely immune (which may be why it does not cause problems in its native home) but well-tended planted trees have thin bark and a high sap content and are mercilessly attacked. So big is the problem that ash, lime and wild cherry may become more common in Britain because of their relatively low palatability to squirrels at the expense of palatable beech and sycamore.

What is a tree?

Everyone knows what a tree is: a large woody thing that provides shade. Oaks, pines and similarly large majestic trees probably come immediately to mind. A stricter, botanical definition is that a tree is any plant with a self-supporting, perennial woody stem (i.e. living for more than one year). The first question that normally comes back at this point is to ask what then is a shrub? To horticulturalists, a ‘tree’ is defined as having a single stem more than 6 m (20 ft) tall, which branches at some distance above ground, whereas a shrub has multiple stems from the ground and is less than 6 m tall. This is a convenient definition for those writing tree identification books who wish to limit the number of species they must include. In this book, however, shrubs are thought of as being just small trees since they work in exactly the same way as their bigger neighbours. Thus, ‘trees’ cover the towering giants over 100 m through to the little sprawling alpine willows no more than a few centimetres tall.

Some plants can be clearly excluded from the tree definition. Lianas and other climbers are not self-supporting (although some examples are included in this book), and those plants with woody stems that die down to the ground each year, such as asparagus, do not have a perennial woody stem.

The woody skeleton

What makes a tree different from other plants is the trunk (or bole) and branches making up the woody skeleton. The main job of this tough, long-lasting skeleton is to display the leaves up high above other lesser plants in the battle for light. As well as support, though, the trunk and branches have two other important jobs: getting water from the roots to the leaves and moving food around the tree to keep all parts, including the roots, alive. But is the trunk just a large connecting drainpipe that keeps the two ends of the trees apart? In many senses, yes, but its structure allows it to do many other things that no mere drainpipe could do.

Starting from the outside is the outer bark, a waterproof layer, over the inner bark or phloem (Figure 3.1). The phloem is made up of living tissue that transports the sugary sap from the leaves to the rest of the tree. Inside the bark is the cambium, which, as will be shown, is responsible for the tree getting fatter. Inside this again is the wood proper or xylem. Although seemingly ‘solid wood’ it is the part of the tree responsible for carrying water from the roots to the rest of the tree. The water moves upwards through dead empty cells. But wood is not entirely dead.

Like other plants, trees have to engage in sex by proxy, using the wind, water or an animal as an intermediary to get pollen from one tree to another (see Box 5.1). Unlike many other plants, the sheer size of trees raises extra problems of pollination, and eventually seed dispersal, which are solved in ingenious ways. The original trees, the conifers, were (and still are) wind-pollinated. The flowering plants (angiosperms), which include hardwood trees, evolved hand in hand with insects to be, not surprisingly, primarily insect-pollinated. Yet some have reverted back to the old way of wind pollination, and for very good reasons. These are linked to geography: most trees in high latitudes are wind-pollinated, but animal pollination (insects, birds and mammals) becomes more important the closer one gets to the tropics, reaching 95% of trees in the tropics. Figure 5.1 gives an overview of general flower structure.

Animal pollination

Animal pollination is primarily the world of the insect; in the wettest Costa Rican forests, for example, 90% of trees are insect-pollinated. But within insect pollination there are different strategies. Some trees, like magnolias, apples, rowan (Sorbus aucuparia), European spindle (Euonymus europaea), some maples, hawthorns (Crataegus spp.) and a long list of others, go for quantity. They are generalists that spread the pollen on a wide range of flies and beetles in the hope that some will arrive on another flower of the right species.

Perhaps the most striking thing about tree leaves is their tremendous diversity in size. The Arctic–alpine snow willow (Salix nivalis), which grows around the northern hemisphere, can have leaves just 4 mm long on a sprawling ‘tree’ no more than a centimetre high (Figure 2.1). Smaller still, the scale needles of some cypresses are nearer a millimetre long. Among the largest of leaves are those of the foxglove tree (Paulownia tomentosa), which on coppiced trees can be over half a metre in length and width on a stalk another half metre long. Such large sail-like leaves are in great danger of being torn by the wind (as in the traveller's palm, Ravenala madagascariensis; see Figure 2.1) so it is perhaps no surprise that big leaves are usually progressively lobed and divided up into leaflets to form a compound leaf. This can lead to even larger leaves: the Japanese angelica tree (Aralia elata) can have leaves well over a metre in length (Figure 2.1). Many palms have feathery leaves over 3 m long and in the raffia palm (Raphia farinifera) up to 20 m (65 feet) long on a stalk another 4 m long.

The leaves are the main powerhouse of the tree. Combining carbon dioxide from the air with water taken from the soil they photosynthesise, using the sun's energy to produce sugars and oxygen. These sugars (usually exported from the leaf as sucrose, the sugar we buy in packets) are the real food of a tree.

Why write a book on trees? The motivation for me came from the frustration of trying to teach a subject where much is known but is scattered over a huge range of journals and books from many countries. There are so many fascinating stories to be told about the ways in which trees cope with the world and the problems of being large and long-lived: they are extremely well designed. Someone had to write this book!

My goal was to draw together strands of information to create a readable book that would answer common questions about trees, set right a number of myths and open up the remarkable world of how trees work, grow, reproduce and die. It is for you, the reader, to judge whether I have been successful. Please let me know where you find errors or would wish to argue with the logic.

I am indebted to all those who helped with this book, especially Roger Davidson and Bill Williams who read and commented on the whole manuscript and numerous colleagues who commented on parts. Val Brown, P.B. Tomlinson, Colin Black and K.J. Niklas kindly provided detailed information. Maria Murphy and Lynn Davy at Cambridge University Press are to be congratulated for their extreme patience with a faltering author. And, of course, my wife and sons are thanked for putting up with my pet project.