Constant species

Alopecurus myosuroides, Bilderykia convolvolus, Elymus repens, Polygonum aviculare, Stellaria media, Veronica persica.

Physiognomy

The Alopecuro-Chamomilletum is the most distinctive weed assemblage in which the annual grass Alopecurus myosuroides is found. Its tufts can be very abundant here, along with Elymus repens, Polygonum aviculare and Stellaria media, each of which can also show locally high cover. Veronica persica and Bilderdykia convolvulus are also constant, though generally in smaller amounts.

The characteristic mayweed of this vegetation, Chamomilla recutita, is only occasional overall but Matricaria perforata is frequent, along with Chenopodium album, Capsella bursa-pastoris, Sinapis arvensis, Sonchus asper, Senecio vulgaris, Avena fatua, Poa annua, Plantago major, Polygonum persicaria and Cirsium arvense. Among the more common occasional are Convolvulus arvensis, Euphorbia helioscopa, Solanum nigrum, Urtica urens and the scarce goosefoot Chenopodium hybridum.

Habitat

This assemblage is typical of winter-sown cereals and other field crops on loamy and clayey soils that occur in the warmer and drier south-east of England but which are protected from drought by their heavy texture.

A. myosuroides is native across much of Europe but is only common in Britain in the more Continental parts of the country where the July mean is above 15 °C (Naylor 1972), being especially abundant in the southeast Midlands and East Anglia. It can get a head start among arable crops in this region by autumn germination and survive through the winter as a rosette, flowering from May to August of the following year.

Synonymy

Sandscale Salix repens dunes Pear sall 1934; Campylium stellatum-Salix repens nodum Jones 1992; Carex flacca-Thalloid Liverwort nodum Jones 1992 p.p.

Constant species

Agrostis stolonifera, Carex flacca, Epipactis palustris, Equisetum variegatum, Hydrocotyle vulgaris, Mentha aquatica, Salix repens, Calliergon cuspidatum, Campylium stellatum.

Rare species

Dactylorhiza majalis ssp. praetermissa and ssp. purpurella, Juncus acutus, Liparis loeselii, Pyrola rotundifolia.

Physiognomy

The Salix repens-Campylium stellatum community comprises generally closed vegetation of dune slacks with an often extensive low bushy carpet of S. repens, frequently species-rich assemblages of vascular associates and a usually extensive carpet of bryophytes among which both Campylium stellatum and Calliergon cuspidatum figure very commonly, the former generally especially abundant.

The constant combination of Carex flacca, Equisetum variegatum and, in most sub-communities, Epipactis palustris with more widely occurring dune-slack plants such as Hydrocotyle vulgaris, Agrostis stolonifera and Mentha aquatica, is what gives this vegetation its particular character. Among these species, E. variegatum can be especially abundant and it sometimes assumes a striking tussock growth-form, perhaps because of rapid shoot proliferation in the early stages of colonisation (Jones 1992). Also, as with the Salix-Calliergon community, the only occasional occurrence of Potentilla anserina and Carex nigra here, and the generally low cover of the latter, is an important contrast with the Potentilla-Carex slack vegetation.

Among the other commoner associates, Carex arenaria, Leontodon autumnalis, Ranunculus flammula, Rubus caesius, Lotus corniculatus and Juncus articulatus are the most frequent with more occasional Taraxacum officinale agg. and Equisetum palustre.

Constant species

Dactylis glomerata, Lolium perenne, Plantago lanceolata, Taraxacum officinale agg.

Physiognomy

The Lolium perenne-Dactylis glomerata community comprises coarse weedy grassland vegetation in which Lolium perenne and Dactylis glomerata usually make up the bulk of the more or less closed cover, along with a variety of perennial associates and scattered ephemerals which find a place in locally disturbed places. Plantago lanceolata and Taraxacum officinale agg. are the commonest of these companions but Achillea millefolium, Plantago major, Trifolium pratense, Agrostis stolonifera, Urtica dioica, Hypochoeris radicata and Potentilla reptans all figure occasionally among the perennials, Poa annua, Bromus hordeaceus ssp. hordeaceus and B. sterilis among the annuals.

Sub-communities

Typical sub-community. Apart from the community constants, only frequent records for Trifolium dubium and Hordeum murinum with occasional Vicia sativa and Senecio squalidus are distinctive here.

Crepis vesicaria-Rumex obtusifolius sub-community. Poa annua becomes constant here but better preferential are Crepis vesicaria and Rumex obtusifolius with Poa trivialis, Senecio vulgaris and Cirsium arvense occasional. Seedlings of Buddleja davidii are sometimes found.

Plantago major-Trifolium repens sub-community. Poa annua and Holcus lanatus remain very frequent here but Plantago major and Trifolium repens are more distinctive with Ranunculus repens and Rumex crispus occasional.

Arrhenatherum elatius-Medicago lupulina sub-community. The grass contingent of the vegetation is further augmented here by constant H. lanatus and, more preferential, Arrhenatherum elatius and Agrostis capillaris.

Synonymy

Includes Brassicetum oleraceae Géhu 1962.

Constant species

Festuca rubra, Brassica oleracea, Dactylis glomerata, Daucus carota ssp. gummifer.

Rare species

Brassica oleracea, Ophrys sphegodes, Silene nutans.

Physiognomy

The Brassica oleracea community generally has an irregular grassy cover of Festuca rubra and some Dactylis glomerata with prominent erect or decumbent plants of B. oleracea and a little Daucus carota spp. gummifer Plantago lanceolata is the most frequent associate throughout, but it is never abundant. Cheiranthus cheiri and Sonchus oleraceus, though infrequent, may be conspicuous when flowering.

Sub-communities

Beta vulgaris ssp. maritima sub-community: Brassicetum oleraceae Géhu 1962. B. oleracea is more abundant in this species-poor sub-community and Beta vulgaris ssp. maritima is an additional constant which may be codominant with B. oleracea and Festuca rubra. Maritime species, such as Armeria maritima, Silene vulgaris ssp. maritima and Bromus hordeaceus ssp. ferronii are confined to this sub-community though none is ever abundant.

Ononis repens sub-community. Additional constants here are O. repens, Silene nutans, Centaurea scabiosa and Rumex acetosa. B. oleracea is less abundant here than in the Beta sub-community. Also frequent are Brachypodium pinnatum (which often co-dominates with F. rubra), Hieracium pilosella and Teucrium scorodonia and among the occasional species are some characteristic of calcicolous grasslands including Ophyrs sphegodes (Summerhayes 1968).

Andrews (1916) eloquently expressed the idea that soil fertility was critical in delimiting rainforest when he wrote that around Sydney:

Subsequently, a number of ecologists have also advocated the idea that soil fertility is critical in determining the distribution of rainforest vegetation (McLuckie and Petrie 1927; Davis 1936; Fraser and Vickery 1938, Pidgeon 1940; Herbert 1960; Florence 1964, 1965; Webb 1968). The object of this chapter is to show that Australian rainforests are not consistently limited by soil texture, drainage, or soil fertility.

Geology and soil types

While it is true that the largest tracts of rainforest in Australia are, and were before being cleared, associated with fertile basaltic soils, the importance of soil parent material in controlling rainforest has been over-emphasised.

In his essay on Australian phytography, Hooker (1860) demonstrated that the Australian vascular flora was not taxonomically unique. Instead, he concluded that the ‘peculiarities of the Flora, great though they be, are found to be more apparent than real’. However, Hooker felt that many components of the Australian flora had a ‘very peculiar habit or physiognomy, giving in some cases a character to the forest scenery (as Eucalypti, Acaciae, Proteaceae, Casuarinae, Coniferae)’ with some plants having ‘an anomalous or grotesque appearance’ (Figure 3.1). The idea that the dominant Australian vegetation has a peculiar physiognomy including a particular prevalence of sclerophylly is central to a number of definitions used to distinguish rainforest from characteristically Australian vegetation (Francis 1951; Beadle and Costin 1952; Webb 1959, 1968, 1978). Webb differentiated non-rainforest vegetation from rainforest on the basis of sclerophylly. Confusingly, he conceded that rainforest trees species could have leaves that are anatomically sclerophyllous (Figure 3.2). He attempted to explain this contradiction by arguing that ‘sclerophylly has no general anatomical justification when used in its ecological sense’ although he suggested that ‘sclerophyll’ non-rainforest species might be physiologically and chemically distinct from rainforest plants that are also anatomically sclerophyllous. Other workers have argued that sclerophylly is essentially a physiological condition that enables non-rainforest plants to grow on infertile soils (Loveless 1961; Beadle 1966; Johnson and Briggs 1981). It is remarkable that despite the fact that there is no consensus as to the meaning of the term ‘sclerophylly’ it has become a central concept in Australian vegetation science.

Rainforests in Australia occur as scattered islands within vast tracts of the quintessential Australian vegetation dominated by Eucalyptus and Acacia. Rainforest ‘islands’ are found from Tasmania in the temperate zone to the monsoon tropics in the north across wide gradients of rainfall, altitude and temperature. In this book, I argue that their patchy distribution is the consequence of tens of millions of years of fires started by lightning and other natural causes. Fire became a feature of the Australian landscape as the continent became progressively drier. This occurred because of the dominance of the subtropical high pressure system beginning in the mid Tertiary when the continent drifted into the mid latitudes, the thermal gradient between the equator and the South Pole intensified, and the Antarctic ice sheet formed. An increase in the frequency of wildfires triggered a major diversification of the fire-tolerant biota. In plants, there was selection for new traits such as serotiny, and the refinement of pre-existing traits such as the lignotuber. Some fire-tolerant vegetation became fire-adapted and ultimately fire-promoting, sealing the fate of most rainforest vegetation. This evolutionary divergence is responsible for the fundamental and often sharp ecological and floristic dichotomy of the current Australian biota into rainforest and non-rainforest types. Thus the evolution of the current Australian flora was forged in a fiery environment in which the ancestral rainforests become scattered and burnt.

Although there is an inextricable link between fire management and rainforest conservation, there is much contention about which fire regimes are appropriate. It is certainly the case that ‘one size does not Wt all’. Even similar rainforest types do not necessarily have the same fire management requirements, because the history of land use can be of overarching importance. For example, in spite of the ecological similarity of Northern Territory and Queensland monsoon rainforests, Russell-Smith and Bowman (1992) and Fensham (1996) came to very different conclusions about the consequences of current fire regimes. Russell-Smith and Bowman found that about one-third of the 1220 rainforest sites surveyed in the Northern Territory had boundaries severely degraded by fire, and concluded that this was the result of the near-complete breakdown of traditional Aboriginal fire management. In contrast, Fensham found that less than 10% of the 358 monsoon rainforest patches he surveyed in inland north Queensland were degraded by wildfire, and concluded that fire is not a major management issue. In north Queensland, intensive pastoralism has reduced grass fuels to such low levels that wildfire has become an unusual occurrence. However, Fensham noted that the situation in Queensland would change if pastoralists began to use fire to control native and exotic woody ‘weeds’ that have established on overgrazed rangelands.

In addition to directly threatening the biological integrity of many Australian rainforests, weeds also have the capacity to substantially change fuel characteristics across rainforest boundaries.

The existence of patches of rainforest embedded in tracts of Eucalyptus forests have long perplexed and sometimes astonished field biologists. The abrupt rainforest boundaries often rise up like ‘a dark wall’ (Figure 1.1) in the relatively open Eucalyptus forest (Herbert 1932) and literally confront ecologists with the question ‘what determines the position of the boundaries?’ The floristic differences between rainforest and Eucalyptus forests can be ‘so great as to suggest separate geographic and historical origins in spite of their growing side by side’ (Herbert 1932). The purpose of this book is to investigate the deceptively simple question of why rainforests have such limited and fragmentary coverage in Australia (Figure 1.2). This basic geographic question raises other questions such as:

1. (i) Why should Eucalyptus and Acacia dominate the great bulk of Australian woody vegetation (Figure 1.3)?

2. (ii) How should rainforest be defined in Australia?

3. (iii) What environmental factors control the local extent of rainforest?

As will become apparent in this book, these questions are central issues in Australian vegetation science. The rainforest-boundary question occurs in nearly all the major arguments about Australian woody vegetation. Indeed, a number of important theories about the ecology of Australian woody vegetation have explicitly sought to resolve this problem of sharply contrasting forest types growing side by side. I recognise that the Australian rainforest boundary problem is a subset of the global biogeographic question concerning the cause of the differentiation between forest and savanna.

The realisation that fire is important in determining rainforest boundaries has naturally led many authors to speculate on the effect of Aboriginal landscape burning. This line of thinking has a long history. Domin (1911) asserted that ‘secondary’ Eucalyptus forests had replaced rainforest in southern Queensland and that this change was due to ‘aboriginal inhabitants, mostly by means of bushfires’. Currently, two temporal perspectives dominate this topic. The first focuses on the possible ecological and evolutionary consequences of the arrival in Australia of a new ignition source: Homo sapiens. Some authors have suggested that a sharp increase in the frequency of fires caused the contraction of rainforest vegetation and possibly stimulated the evolution of fire-adapted Australian vegetation during the last ice age. For instance, Truswell (1993) concluded a review of the Cainozoic history of Australian vegetation with the thought that ‘the expansion of the eucalypts, probably at the expense of the drier rainforests and Casuarinaceae woodlands, is a more recent story and has possible links with human land-management practices’. The second perspective concerns the more recent past, and argues that landscape burning by Aborigines had controlled the distribution of rainforest throughout Australia. This view is largely the product of ecological studies of rainforest boundary dynamics. Several authors consider fire as being an environmental constant through geological time (Mount 1979; Horton 1982). They see Aboriginal landscape burning as being irrelevant in determining current and past vegetation patterns.

Webb (1968) noted that although rainforests typically favour nutrient-rich or eutrophic soils, this correlation is not causal in the sense that rainforest trees have a physiological dependence upon high levels of soil nutrients. Rather, he argued that the occurrence of rainforest on eutrophic soils is more a consequence of the overarching effect of fire on the growth and survival of rainforest trees. Thus soil fertility is not merely determined by the underlying geology of a site but by fire history, nutrient cycling and topography. For instance, Webb argued that, in seasonally dry and geologically infertile environments, rainforest is restricted to topographically fire-protected sites, and that the absence of burning allows rainforest to enrich soils by nutrient cycling. The purpose of this chapter is to critically consider the role of nutrient cycling, fire and topography in creating differences in soil fertility between rainforest and non-rainforest vegetation.

Nutrient cycling and rainforest soil fertility

The view that differences in nutrient cycling contribute to the higher soil fertility of rainforest compared with adjacent non-rainforest is supported by the available data. For instance, Chandler and Lamb (1986) provide data on differences in nutrient use by rainforest and non-rainforest vegetation using the index of ‘nutrient use efficiency’ developed by Vitousek (1984).

In an insightful paper, Andrews (1916) suggested that the dominant Australian flora was originally derived from rainforest and had undergone extreme morphological modifications in order to cope with infertile soils. This view was subsequently elaborated by the ecologist Beadle (1953, 1954, 1962a,b, 1966, 1968), who argued that the restricted distribution of Australian rainforests is caused by the widespread exhaustion of soil phosphorus through geological time. Beadle (1966) took this view to the extreme by stating that ‘the absence of rainforest and the paucity of rainforest genera in south-west Western Australia are accounted for by low [soil] fertility levels rather than by a decimation of rainforest genera by a past dry climate’. As discussed in the previous chapter, Beadle argued that the non-rainforest ‘sclerophyll’ vegetation owed its dominance to physiological characteristics that enabled it to survive on infertile soils, particularly those with extremely low levels of phosphorus.

Beadle (1953, 1954, 1962a, 1966, 1968) supported his hypothesis by examination of correlations and some experimentation. Although his hypothesis has received general support (Ashton 1976a; Attiwill et al. 1978; Webb 1969) and only limited criticism (Coaldrake and Haydock 1958; Le Brocque and Buckney 1994), there have been few tests of the hypothesis (Coaldrake and Haydock 1958; Richards 1968; Adam et al. 1989; Le Brocque and Buckney 1994). The purpose of this chapter is to demonstrate that available data do not support the generalisation that rainforest is restricted to soils with high soil phosphorus levels.

Schimper's (1903) original use of the term ‘rainforest’ succinctly encapsulated his belief that luxuriant, Xoristically and structurally complex tropical vegetation could only develop in a drought-free climate. Although some Australian ecologists have adhered to the idea that moisture supply is of critical importance in controlling the distribution of Australian rainforest (Specht 1981a), there are several features of Australian rainforests that are inconsistent with this theory. Firstly, there is an imperfect correlation between high rainfall areas and the distribution of rainforest, a fact that puzzled pioneering ecologists such as McLuckie and Petrie (1927). Secondly, in the latter half of this century many Australian ecologists broadened the concept of rainforest to include vegetation that occurs in low rainfall and seasonally dry climates (Webb 1959; Baur 1968; Webb and Tracey 1981; Adam1992) (Figure 6.1). This broadening of definition is reflected in the use of the oxymoron ‘dry rainforest’ (Baur 1957; Adam 1992; Fensham1995). Indeed, many authors have opted to spell rainforest as a single word (rather than the more conventional ‘rain forest’) in order to play down the implication that high rainfall is critical in controlling the distribution of rainforest in Australia (Baur 1968; Webb and Tracey 1981; Adam1992). Thirdly, there is evidence that some Australian rainforest trees, including some that occur in high rainfall environments, are extremely drought tolerant, being able to endure seasonal moisture deficits with ‘leaf water potentials more commonly associated with arid environments’ (Yates et al. 1988).

In 1959, two papers that laid the foundations for the modern perspective on Australian rainforest ecology were published. Webb (1959) published the first systematic classification of Australian rainforest vegetation across its geographic range from Tasmania to the monsoon tropics. His project highlighted the enormous variability of Australian rainforest, notwithstanding that he excluded the transitional or ecotonal vegetation that often separates rainforests from sclerophyll vegetation from consideration. Coincidentally, Gilbert (1959) explored the dynamics of a peculiar Tasmanian forest type that was transitional between Nothofagus cunninghamii rainforest and tall Eucalyptus regnans forest. Gilbert's study was important in crystallising our perception of the importance of fire in Australian forest ecology in general and in the dynamics of Nothofagus rainforest boundaries in particular.

A decade later, two influential papers advanced this early work. Webb (1968) considered the environmental forces that limit the distribution of Australian rainforest. Although emphasizing a multiplicity of forces, Webb identified the fundamental importance of fire in influencing the distribution of rainforest throughout its range. In the same year, Jackson (1968) integrated a large number of ideas about the dynamic response of western Tasmanian vegetation, including Nothofagus rainforest, to landscape fires. Two important concepts were developed in this paper: the evolutionary relationship between fire and vegetation, and the effect on secondary successions of the variation in the interval between successive fires.

In 1969, Jones published a paper which articulated the view that Aborigines had intentionally modified their environment, including rainforest, using fire.

In this book, I have sought to establish that the boundaries between rainforest and non-rainforest vegetation are determined by the frequency of wildfire. This generalisation holds across a latitudinal gradient from12° S in the monsoon tropics to 44° S in Tasmania and includes all rainforest environments on the east coast of Australia. Indeed, in arid Australia fire also delimits the boundaries of small patches of woody vegetation that have taxonomic links to ‘rainforest’, such as Callitris-dominated communities (Bowman 1992b; Bowman and Latz 1993). The importance of fire in dichotomising the ecology of Australian forests was clearly enunciated by Webb (1968). He wrote that it is possible to ‘regard the Australian forest flora as composed of species ranging from extremely fire-sensitive to fire tolerant, which have been sifted by fire as an evolutionary factor to produce two major classes of vegetation: one whose existence depends on protection from fire, and the other which is able to survive or regenerate in different degrees after burning’. The purpose of this chapter is to explore the historical reasons for this ecological divergence. First, I will briefly describe the environmental history of Australian vegetation by reference to the abiotic and biotic evidence. I will then critically consider evolutionary models that account for the diaspora of the habitat islands of fire-sensitive rainforest in vast tracts of flammable forests.

The Europeans who colonised Australia had no conceptual frame with which to deal with the vast tracts of evergreen native vegetation that they came to, at Wrst disparagingly, then affectionately, call ‘bush’. The place name ‘Botany Bay’ attests to the impact that the ‘strange’ Australian flora had on the earlier explorers. In the nineteenth century, the dominant Australian vegetation aroused emotions of alienation, fear, despair, desolation, and melancholia in many observers. For example, the author Marcus Clarke wrote in 1876 that ‘The Australian forests are funereal, secret, stern. Their solitude is desolation … No tender sentiment is nourished in their shade …’ (Ritchie 1989). One exception was the luxuriant, cool and shaded forests that formed sanctuaries in the vast tracts of dry ‘monotonous’ bush (Ritchie 1989). In the nineteenth century these moist forests were known, not as ‘rainforests’ but by a range of terms including fern-forest, brush, scrub, and jungle (Adam 1992). In the twentieth century, the term ‘rainforest’ has been almost universally used and understood to denote an atypical Australian forest type. Sadly, we have limited appreciation of Aboriginal knowledge or terminology of Australian vegetation: only a few Aboriginal terms for Australian plant communities, such as ‘mulga’ and ‘mallee’, have been incorporated into the English language.