encourage too many fires, which prevents the spinifex from recovering as an important fuel. Too many dry seasons prevent the eruption of ephemerals that carry fire from one coughing hummock to the next.
Still, spinifex makes an extraordinary fuel. For three years after burning, spinifex sprouts are palatable and nutritious, with up to 6 percent crude protein. Then its dietary value collapses and, with less than 3 percent protein, it becomes all but inedible. Its decomposition becomes almost wholly restricted to fire; virtually all new growth is available fuel. The hummocks do what they can to encourage combustion: live stems are rich in resins that burn fiercely, dead stems are typically dry, and all of the hummock is thoroughly ventilated. Spinifex and its associated flora weave a powerful fuelbed; a mature community may groan under a load of 3–8 tons/hectare in available fuels. Once begun, there is little in the steppe lands that spinifex favors to break down a major conflagration. Only changes in winds or surges of moisture or the past history of fires, as recorded in large-scale fuel mosaics, can modulate a free-burning blaze.
The fire history of spinifex shapes the biodiversity of the entire ecosystem. Spinifex pervades 20 to 80 percent of the total ground cover, and it erects a structural matrix for the remaining flora and fauna. In the Simpson Desert hummock grasslands support 180 species of plants; in the central Australian sandplain, 154 species. Combustion releases species—fire ephemerals—that lie otherwise dormant. In regions of little variability, spinifex’s peculiar growth habits thus provide a diversity of habitats. Its microniches testify to a history of disturbance, which is largely a history of fire. Without fire the hummocks become decadent, the landscape uniform, and the cycling of nutrients feeble. Even the hardest-seed ephemerals spoil.
Remove fire and watch mulga, Callitris, and other fire-sensitive refugees timidly reconquer a site. But, paradoxically, remove reliable rains, even precipitation as meager as that in the spinifex, and watch fire retreat before stony desert or a renewed mulga mélange. Free-burning fire requires a full-flushed biota. What powers fire in the spinifex is the mingling of seasonal rains with seasonal drought. That pattern makes spinifex into a kind of central fire, and a model for the two-cycle engines that drove the centrifugal pump that was Old Australia.
THE WET AND THE DRY
Northward, a hesitant, spotty monsoon—its southern fringe brushing against deserts—hardens with grim finality into summer and winter, the Wet and the Dry. Spinifex grades into savanna, hummock grasses into tropical swards; the Australian tropics ripen into a voluminous grassland studded with trees. Sorghums dominate the north, kangaroo grass (Themeda) the eastern steppes, and black spear grass (Heterogon) the coastal woodlands. Eucalypts are universal. On wetter sites, they share dominance with Melaleuca; in more arid lands, with Acacia. Where they can be shielded from fire, mangrove swamps cling like barnacles to the tidewater streams, enclaves of Callitris blossom, and patchy rainforests endure. But there is little protection from fire.5
The climate of the wet-dry tropics makes routine fire possible; its biology makes it inevitable. Soils are heavily laterized, and even by Australian standards, nutrient-drained. The monsoon and the biochemical cycles it brings to life give the nutrient flow a strongly seasonal dimension—captured into biomass during the Wet, released by fire during the Dry. The expansive grasses focus the action.
Those dominant grasses are large in biomass but small in nutritional content. Their quality, not their quantity, limits their harvesting by consumers. With the onset of the Wet, grasses spring to life and by the end of the growing season yield standing biomass on the average of 375–625 tons/hectare in central Queensland or 227 tons/hectare around Katherine in the Northern Territory, the difference between the Australian llanos and the Australian sahel. But after an early flush, the protein content decays to an abyssmal 2.5–3.0 percent by the end of the season, and grasses typically transfer important nutrients, including organic nitrogen, to underground storage. Few fauna consume the dead stalks. Without removal, without recirculating the precious nutrients, future growth falls off rapidly. Seed regeneration falters, finer grasses supersede the coarse sorghums, and woody shrubs suppress the grasses altogether.
Thus the life cycle and the fire cycle of the tropical grasses converge with machined precision. Burning stimulates biomass production (5–10 percent over that of unburned sites) and enriches its crude protein content by a factor of four or five. Fire brings the biota to life. Whole food chains—from invertebrates to raptors—collect around a moving fire. Grazers rush to the green pick that pokes through the ashes soon afterward. Even termites preferentially invade trees on burned sites rather than members of the same species in rainforest. By the end of the Dry, fire has readied the savannas for new growth, and it has even burned lowlands that, at the height of the Wet, are flooded. Fresh rains act on cleared sites and mobilized nutrients to turn black to green.
Its grasses establish the fire regime. The wetter sites are burned annually; the drier, once every two or three years. The exceedingly low nutrient reservoir and high fire frequency affect everything in the system. Unlike Eucalyptus elsewhere, the tropical eucalypts do not thrive on this fire environment. They survive. They are stunted, marginal; they can barely capture sufficient nutrients to sustain themselves; they coexist in uneasy equilibrium between rainforest and savanna. Although fire rolled back the rainforest sufficiently for eucalypts to transgress into the region, it now promotes the tropical grasslands with a fire frequency that has left the eucalypts living on the margin. While the eucalypts and paperbarks can survive the fires, they cannot compete as aggressively for the liberated nutrients, for annual fires constantly short-circuit the cycles that the trees demand. Even minor changes in fire frequency and the scale of burning can shift the balance of power. Biennial firing leaves the mosaic of woods and grasses stable; annual firing pushes it toward the grasses. The biota is poised on a knife-edge, sensitive to any variation in fire frequency.
This sensitivity makes it difficult to reconstruct the fire history of Old Australia. In the interior, routine fire is impossible because of limited fuels, which reflect uncertain rains. Outside that barren core, however, fuels are generally ample, and the patterns of fire history reflect the patterns of ignition. In Old Australia the early storms of the Wet brought lightning, forking like a lizard’s tongue. The old grasses—once tall, now laid low by monsoon winds—readily kindled. The drier sites burned first. A texture of burns resulted, a mosaic of black soot and yellow grass, as dry, unburned patches took fire. The Australian savanna—like those in tropical Africa and South America—formed a belt between rainforest and desert. As the monsoon transgressed and regressed, that savanna expanded and contracted, marched south and retreated north.
It is difficult to trace exactly these dimensions because, during the Pleistocene, a new ignition source appeared, the Aborigine. From that moment onward human uses, not natural sources, dictated fire frequency. The fire regime of the wet-dry tropics dates from that event as firesticks imposed new fire frequencies and timings onto different portions of the biota. Rainforest retired to special enclaves, more or less deliberately spared from fire. The anomalies that presently exist—such as the status of the eucalypts—date from changes in human fire use that accompanied European settlement. Thus, in this exfoliating geography of Australian fire, the increase in fire from the center outward reflects also an increase in human ignition. Humans sought out fire, added to it, and through it reshaped Old Australia.
THE SOUTHWEST ENCLAVE
In the southwest the pattern of seasonal wet and dry became Mediterranean, a cycle of winter rains followed by a long, thirsty summer. Its biotas are complex, syncretic; they form an easy enclave, the product of a double isolation—the first as Australia departed Gondwana, and the second as encroaching seas and later deserts divided the emerging Australian scleroforest into east and west. Eucalyptus rules the forest; scleromorphs fluff the understory. Toward the interior, woody savanna grades into spinifex, the mulga mélange, and unburned desert. On some sandplains heaths flourish, and on many arid ridges, mallee. Endemism is extraordinarily high, and fire is everywhere.6
