Smoke Germination of Australian Plants

Kingsley Dixon

(This article is reproduced with permission from the Newsletter of the Australian Flora Foundation, January 2017)


Fire has played a significant role in the ecology of the Australian flora at least since the arrival of arid conditions in the mid-Tertiary. In fire-prone floras, particularly those of mediterranean zones, fire has been shown to be crucial for the recruitment from seed of a wide variety of taxa. For seeder or fire-sensitive species and fire ephemerals, fire is the single most important cue for triggering germination of the dormant soil seed bank.

For many fire-responsive taxa, germination of viable seed under controlled conditions has been difficult or impossible using conventional treatments other than in vitro excised embryo culture or complex and often difficult to apply pre-treatments including hormonal applications. However, my research has now shown that smoke responsive species occur throughout the world's plants from fire-prone to non-fire-prone floras.

The role of smoke in germination

Smoke is a key principal in breaking seed dormancy in a wide variety of native Australian species. Many studies over the years since the discovery for Australian species in 1993 of the action of smoke in germination have found that smoke:

  • Promotes faster and more uniform germination under controlled greenhouse and laboratory conditions.
  • Enables germination in species previously thought difficult or impossible to germinate by conventional means e.g. Geleznowia and Eriostemon (Rutaceae); Hibbertia (Dilleniaceae); Stirlingia and Conospermum, Grevillea and Hakea (Proteaceae); Verticordia and Calytrix (Myrtaceae); Pimelea (Thymeleaceae); Blancoa (Haemodoraceae); Stylidium (Stylidiaceae).
  • Promotes germination in species withlow levels of germination e.g. Anigozanthos and Conostylis (Haemodoraceae); Thysanotus and Burchardia (Liliaceae); Patersonia (Iridaceae); Lechenaultia (Goodeniaceae); Gyrostemon and Codonocarpus (Gyrostemonaceae); Stackhousia (Stackhousiaceae); Hybanthus (Violaceae).
  • The promotive effect is independent of seed size and shape; plant life form i.e. whether annual, perennial, herbaceous, seeder (fire sensitive) or resprouter (fire-tolerant).
  • Aerosol smoke, smoke dissolved in water or direct smoked solids (activated clays, sand particles) or direct-smoked seeds are effective methods for delivery of smoke for seed germination.
  • High doses of smoked water can inhibit germination of many species e.g. paper daisies (Asteraceae).
  • As seed ages it can change in its ability to respond to smoke such as in Conostylis which has cyclical dormancy - seed dormancy cycles in and out of smoke responsiveness each 6 months or so.
  • Ageing seed in soil for 1-2 years and repeat smoke application will result in high germination of smoke 'recalcitrant' species in the Ericaceae, Proteaceae, Cyperaceae, Restionaceae, Rutaceae and others.

Smoke methods (see figure below)

Sown seed trays or seed are placed on an open mesh, two tiered frame in a sealed, plastic tent approximately 2 x 2 m and 1.4 m high. Smoke is generated by slow, controlled combustion of a mixture of fresh and dry leaf and twig material from a range of plants avoiding too much myrtaceous material due to the high level of oils in the leaf material that can be 'distilled' over onto the seed trays. The drum is fitted with an inlet through which air is pumped at the rate of 60-100 litres per minute, and an outlet connected to a 1.5 m long pipe. A 2 m length of flexible stainless steel exhaust piping approximately 50 mm in diameter is connected to the plastic enclosure ensuring that smoke is injected towards the roof of the tent. This ensures that there is adequate spread of smoke inside the tent.

After smoking for 60 minutes, trays are transferred to the glasshouse and watered carefully for the first 6-10 days to ensure that the soluble promoter in smoke comes in contact with the seeds but not washed through the mix before reacting with the seed. Watering is then continued as for normal germination.

Seeds can also be direct smoked. In this instance, air-dried seed is laid out in a single layer in trays. The trays are smoked for 60 minutes in the fumigation tent (as described above) and seed is then sown or stored dry until required. Unlike smoke applied to soil containing sown seeds, smoked seeds can watered as for normal seed trays.

Smoke Germination
Fig 1. Schematic of how to use smoke for germination of native Australian seeds

Smoke water

Smoke water can be useful for direct priming or pre-germination of seeds prior to sowing. Smoke water-treated seeds have the advantage of not requiring the use of the smoke tent and the convenience of priming seeds at will. Smoke water-treated seeds may germinate better than smoked seedling trays with the process applicable to handling potentially large quantities of seed such as for land restoration or automated seed sowing devices.

Smoke water is produced by drawing smoke produced from the combustion drum operating as for aerosol smoke, through a 20 litre container of water. Smoke bubbling is done for approximately 60 minutes and the resultant solution is frozen till required. It is important to note that commercial smoke water products are available but if they are derived from wood this smoke water can be highly suppressive of germination.

Seed to be treated with smoked water is soaked for 12 hours in a 10% solution of the neat solution and then sown or dried then sown as required. Seed treated with smoked water can be watered normally after smoking. Although this method has been shown to be useful for a number of native species, caution is recommended as seed of some species can degenerate if soaked in water for prolonged periods. Pre-germination as a horticultural practise for seed of Australian native plants requires some experimentation to ensure the process is applicable.

In situ habitat germination studies

Smoke fumigation treatments can be applied directly to habitat sites and for a range of species germination will happen in 6-8 weeks after treatment. This technique is particularly useful for soil seed bank auditing (determining how large a native soil seed bank is) and for patch stimulation of rare species. The benefit is that the process is defined rather than the use of prescribed fire which can result in highly patchy germination.

Smoke is generated as above and applied to sites where excess leaf litter and larger plants have been removed to prevent 'shadowing' of the soil from smoke. Tents are erected over the sites, usually 5 x 1 m and 40 cm high, and smoke is pumped into the tent for 60 minutes. In temperate regions of Australia, the best results are achieved if smoking is done in summer to early autumn so that leaching of the smoke products coincides with the onset of the first rains. Smoking undertaken at other times of the year appears to yield less germination for taxa which respond to summer/autumn smoking.


Smoke derived from the combustion of plant material enhances seed germination in a wide range of native species that will not germinate under normal nursery conditions. Species in the families Rutaceae, Restionaceae, Ericaceae, Thymeleaceae, Proteaceae and Dilleniaceae and many other species in other families have improved germination in response to the application of smoke, smoke water or the key germination promoting agent in smoke, karrikinolide. Genera known to be highly recalcitrant to conventional seed propagation which respond to smoke include Stylidium (Stylidiaceae), Geleznowia (Rutaceae), Hibbertia (Dilleniaceae), Stirlingia (Proteaceae), Verticordia (Myrtaceae), Actinostrobus (Cupressaceae) and Pimelea (Thymeleaceae). Germination percentages of species which normally germinate in small numbers are positively influenced by smoke treatment. Seed size varied amongst all positively responding taxa. Within the Ericaceae, small-seeded (Lysinema and Sphenotoma) but not larger, woody-fruited species respond to smoke however both seed types respond under field stored seed. Significant applications now exist for the use of smoke in germinating a wide range of species for horticulture and land restoration.


Bell DT, Plummer JA, Taylor SK (1993) Seed germination ecology in southwestern Western Australia. Botanical Review 59, 24-73.

Brown NAC (1993) Promotion of germination of fynbos seeds by plant derived smoke. New Phytologist 122, 1-9.

de Lange JH, Boucher C (1990) Autecological studies on Audouinia capitata (Bruniaceae). I. Plant-derived smoke as a seed germination cue. South African Journal of Botany 56, 700-703.

Dixon KW, Roche, Pate JS (1995) The promotive effect of smoke derived from burnt native vegetation on seed germination of Western Australian plants. Oecologia 101, 185-192.

Kemp EM (1981) Pre-Quaternary fire in Australia. In: Fire and the Australian Biota, Gill AM, Groves RH, Noble IR (eds.) Australian Academy of Science, Canberra, 3-21.

Meney KA, Neilssen GM, Dixon KW (1994) Seed bank patterns in Restionaceae and Epacridaceae after wildfire in kwongan in southwestern Australia. Journal of Vegetation Science 5, 5-12.

About the author

Professor Kingsley Dixon is an academic in the Department of Environment and Agriculture at Curtin University and at the University of Western Australia and is a Visiting Professor at Kings Park and Botanic Garden. In addition, he is a valued member of the Scientific Committee of the Australian Flora Foundation. His research areas are in conservation science, restoration ecology and plant science. Professor Dixon was named Western Australian Scientist of the Year in 2016.

From the newsletter of the Australian Flora Foundation, February 2017.

◄◄ Articles Index    ◄ Propagation from Seed    Top ▲