Cycas armstrongii

Cycas armstrongii is a cycad of Australia’s Northern Territory that grows in the eucalypt savanna woodlands around Darwin, where it forms dense and very extensive populations. Described in 1868 by Friedrich Anton Wilhelm Miquel and then included for more than a century in Cycas media, this member of the genus Cycas behaves in a way that is very unusual among cycads: in its monsoonal homeland it often drops all its leaves at the end of the dry season. It is the smallest of the Australian Cycas, and the IUCN classes it as Vulnerable (2022).

Key facts about Cycas armstrongii

  • Accepted name: Cycas armstrongii Miq. (1868), accepted by The World List of Cycads (version 2026.08.15) and by the Flora of Australia; POWO could not be consulted directly
  • Former names and synonyms: Cycas media var. inermis A.DC. (1868); included in Cycas media from Bentham (1873) until 1980 (see Taxonomy)
  • Family: Cycadaceae, order Cycadales
  • Common names: fire fern, a local Northern Territory name for the new fronds that this species and Cycas calcicola put up after fire (PACSOA, 1990); the Royal Botanic Garden Sydney labels it “Zamia palm, fire fern” (see Common names)
  • Natural range: Australia, Northern Territory: around Darwin, from the Adelaide River west to the Finniss River and south to the township of Adelaide River, with scattered occurrences almost as far as Hayes Creek and in a small area of Melville Island
  • IUCN status: Vulnerable (VU A4cde), assessed in 2022; also listed as Vulnerable under Northern Territory law
  • CITES and other protection: CITES Appendix II (seeds exempt under annotation #4); Annex B in the European Union
  • Documented hardiness: no data (see Cold hardiness)
  • Light and soil: open forest of Eucalyptus miniata and Eucalyptus tetrodonta, on sands over Tertiary laterite, in full sun
  • Habit and adult size: trunk up to 3 m, rarely 6 m, 5 to 11 cm across; leaves 55 to 90 cm long
  • In the UK, US and Australia: listed by Montgomery Botanical Center (Florida) and the Royal Botanic Garden Sydney in their online catalogues; no record in British gardens (see In cultivation in the UK, US and Australia)
  • Toxicity: no analysis specific to the species; the seeds were eaten by Aboriginal people only after long preparation; treat the plant as poisonous (see Toxicity and precautions)

How to recognise Cycas armstrongii

It is recognised by its new flushes, which turn glossy green, by its two-toned leaflets, with a midrib that stands out equally on both surfaces, and by its size, the smallest among the Australian Cycas, according to Ken Hill (1996) and the Flora of Australia. Hill’s description:

  • Trunk: up to 3 m, rarely 6 m, slender (5 to 11 cm in diameter); often a large underground caudex.
  • Leaves: 55 to 90 cm long, usually slightly keeled (the two rows of leaflets form an angle of 130 to 160°, sometimes up to 180°), with 100 to 220 leaflets, often ending in a spine; petiole 10 to 25 cm, occasionally up to 35 cm, usually hairless and partly spiny.
  • Median leaflets: 5.5 to 14 cm long and 4.5 to 8 mm wide, set at 70 to 90° to the rachis and 6 to 14 mm apart, narrowed to 3.5 to 5 mm at the base; dull green or slightly bluish when they emerge, then glossy mid-green; flat, with flat margins, paler beneath; midrib raised to about the same extent above and below.
  • New growth: covered in loose white down with a few orange hairs, soon shed.
  • Cataphylls: linear, pungent, 6 to 9 cm long (Flora of Australia), with dense orange felt.
  • Male cones: ovoid, orange, 11 to 20 cm long and 7.5 to 10 cm wide; the terminal spine of the microsporophylls is slender, upturned, 7 to 10 mm long.
  • Megasporophylls: 13 to 22 cm long, with grey and orange down, bearing 2 to 4 ovules; sterile blade triangular to lanceolate, 30 to 70 mm long and 18 to 35 mm wide, with 20 to 28 lateral teeth; terminal spine 10 to 25 mm.
  • Seeds: flattened-ovoid, green turning orange, without a waxy bloom, 34 to 37 mm long and 32 to 36 mm wide.

Miquel separated the species by its spineless petiole. Hill showed that this character is unreliable: the petiole is often spiny at least in part, and it is entirely unarmed in only about 10 to 15 % of the plants in a population.

Hybrids

Ken Hill (1996) reports wild populations that are intermediate with Cycas conferta (at Mount Bundey and near Hayes Creek) and apparent hybrids with Cycas maconochiei subsp. maconochiei. A genetic study published in Annals of Botany (Clugston and colleagues, online in 2025, 2026 volume) analysed 236 plants from 26 populations, including one population thought to be a hybrid between the two species. It found no clear signal of hybridisation, but only minimal genetic differentiation between the two taxa (see Taxonomy). No garden hybrid has been documented.

Possible confusion

Several closely related Cycas grow in the Darwin region:

  • Cycas conferta, the species with which it has most often been confused: longer petioles that are almost always unarmed (in about 90 % of plants), dull pale green, slightly bluish leaflets, crowded and overlapping (5 to 8 mm apart) and alike on both surfaces; a larger plant overall (trunk up to 4 m, 9 to 13 cm across);
  • Cycas maconochiei subsp. maconochiei, between Fog Bay and Port Darwin: leaflets with slightly recurved margins, grey or orange-brown felted and glaucous as they emerge, midrib more prominent beneath, larger male cones (21 to 34 cm). The 2025–2026 genetic study suggests that the two taxa probably form a single species with variable morphology;
  • the species of series Calcicolosae (Cycas calcicola, Cycas arnhemica), further south and east, which have soft cataphylls and leaflets with recurved or revolute margins;
  • Cycas media, from Queensland, in which the species was included for over a century: longer leaflets (11 to 28 cm long and 6 to 9 mm wide according to de Laubenfels and Adema) and a range far away.

Taxonomy

The World List of Cycads accepts Cycas armstrongii; POWO could not be consulted directly. Miquel described it in 1868 in the Archives Néerlandaises des Sciences Exactes et Naturelles (volume 3, pages 235–236). In the same paper he cited in synonymy a form inermis of Cycas media that he had discussed in earlier works without ever publishing it validly; in the same year Alphonse de Candolle published Cycas media var. inermis, now a synonym (Hill, 1996).

In 1873 George Bentham sank Cycas armstrongii into Cycas media, and in 1932 Julius Schuster still placed it in the typical variety of Cycas media. It was John Maconochie, botanist to the Northern Territory government, who recognised it again as a distinct species in 1980. In 1994 Ken Hill made it the type species of series Armstrongianosae, defined by hard, pungent cataphylls, flat leaflets with margins that are not recurved, and small male cones; in his 1996 revision this group also includes Cycas conferta, Cycas orientis and Cycas canalis from the Northern Territory, Cycas lane-poolei from Western Australia, Cycas xipholepis from Queensland and Cycas papuana from New Guinea.

David de Laubenfels and Frits Adema (Blumea, 1998) took a much broader view: they included Cycas papuana in Cycas armstrongii and remarked that the latter “could easily be treated as a variety of Cycas media”. Current checklists have not followed them.

Genetics has reopened the question. The RADseq study by James Clugston and colleagues (Annals of Botany 137: 415–430) found low genetic diversity in Cycas armstrongii (expected heterozygosity of 0.038 or less) and minimal differentiation among populations (2.41 % of the variance) and between Cycas armstrongii and Cycas maconochiei subsp. maconochiei (1.81 %). The authors conclude that the two taxa probably represent one morphologically variable species. No formal revision has yet acted on this: for now, The World List of Cycads keeps the two species apart.

Etymology and history

The epithet honours John Armstrong, a collector for Kew who was appointed in 1838 to establish a government garden at Port Essington, on the Cobourg Peninsula, and who later worked in Timor from 1839 to 1845 (Hill, 1996). His dates differ between sources. The World List of Cycads, following Haynes (2022), gives 1820–1902, whereas the biographical notes of the Australian National Herbarium (drawn from Orchard, 1999, and George, 2009) give an unknown date of birth and his death at Kupang, in Timor, on 21 January 1847, with Australian collections from 1838 until at least August 1846. The second version seems more consistent with a career already under way in 1838, but that is our own inference: no published source discusses the discrepancy.

The type specimen (Armstrong 380) is labelled “Port Essington”. Ken Hill points out, however, that the species grows only in the Darwin region, so the type must have come from there, well to the west of the settlement: Armstrong also collected far from Port Essington. Sources disagree on where the holotype is kept. Hill (1996) places it at Kew (K), with a fragment at Utrecht (U); de Laubenfels and Adema (1998) and The World List of Cycads give the holotype at Utrecht, with isotypes at Kew and Paris (P).

Common names

One English name is attested: fire fern. An article by L. P. Butt in the magazine Palms & Cycads (no. 27, 1990), reproduced on the wiki of the Palm and Cycad Societies of Australia (PACSOA), states that the new fronds that both Cycas armstrongii and Cycas calcicola produce after the dry-season fires are known locally as “fire-fern”. The Royal Botanic Garden Sydney labels the species “Zamia palm, fire fern” in its online catalogue. The botanical references (Flora of Australia, Hill’s revision, The World List of Cycads) give no common name, nor any name in an Aboriginal language.

In the wild

According to Hill (1996) and the Flora of Australia, Cycas armstrongii is an abundant species that forms dense, extensive populations on sands over Tertiary laterite, in Eucalyptus miniata and Eucalyptus tetrodonta forest around Darwin. It ranges from the Adelaide River in the east to the Finniss River in the west and south to the township of Adelaide River, with scattered occurrences further south almost to Hayes Creek and in a limited area of Melville Island (Tiwi Islands). Some recent summaries, which rely on Dixon (2004), also mention the Cobourg Peninsula; this could not be checked at the source.

The plant is facultatively deciduous: at the end of the dry season of the monsoon cycle it often loses its leaves, and it is not unusual to see fertile plants carrying only cones. The regular dry-season fires, almost annual over much of the range, also promote leaf fall. They can kill the above-ground parts of the shorter plants, which then branch underground from a caudex that is often greatly swollen (Hill, 1996). According to Hill, several Northern Territory species, this one among them, are probably the most abundant cycads in the world, with populations numbering tens of millions of plants.

The pollinator and the animals that disperse the seeds are not documented.

Conservation status and regulation

Cycas armstrongii is classed as Vulnerable (VU, criterion A4cde) by the IUCN, in the assessment by John Donald Bösenberg published in 2022 (with a correction in 2023), according to The World List of Cycads and the Flora of Australia. Criterion A4 covers a population decline over a period that spans both past and future. In the Northern Territory the species is listed as Vulnerable on the 2012 list under the Territory Parks and Wildlife Conservation Act 1976. Ken Hill did not regard it as threatened in 1996, because of its abundance, but he already noted that overly frequent fires prevent reproduction and that urban development is steadily taking a large part of its habitat. The Red List itself could not be consulted directly.

The genetic study by Clugston and colleagues recommends, besides a new threat assessment, recognising genetically poor or isolated populations (those of the Tiwi Islands, for example) as management units, establishing ex situ collections for those at risk and encouraging gene flow between compatible populations.

Like every species in the genus except Cycas beddomei, Cycas armstrongii is listed in CITES Appendix II, where the family Cycadaceae has been included since 4 February 1977. The listing carries annotation #4 (CoP20 version, in force since 5 March 2026): seeds, pollen, seedlings and tissue cultures obtained in vitro and transported in sterile containers, and cut flowers of artificially propagated plants, are not covered by CITES, whereas international trade in plants requires an export permit. The United Kingdom, the United States and Australia apply CITES through their own national legislation; in the European Union the species is in Annex B of Regulation (EC) No 338/97, whose annexes in force are set by Commission Regulation (EU) 2026/1383, applicable since 29 June 2026. Species+, the official CITES database, confirms this status (consulted on 12 October 2026).

This general information does not replace the official texts: check the rules in force before any purchase or exchange, ask the seller for proof of origin and prefer plants raised from seed in cultivation.

Growing Cycas armstrongii

In short: this is a plant for a heated glasshouse, or for the open garden only in the tropics. No institution has published growing recommendations specific to Cycas armstrongii. The species is grown in gardens in the tropics: a member of the French forum Fous de palmiers living in Thailand photographed it in September 2016 at the George Brown Darwin Botanic Gardens.

The only published observation on its behaviour in cultivation comes from Ken Hill (1996): cultivated plants, like those growing in locally wetter sites, may keep their leaves for longer, but they still go into a resting phase, and strong seasonal stress seems to be needed to bring them back into growth. Its habitat also points to a plant for full sun, a tropical climate with a long dry season, and sandy, free-draining soil.

Outdoor and in-ground cultivation

No result of open-ground cultivation outside the tropics has been published. A note published in 2021 by the French nurseryman Simon Lavaud on the basics of growing Australian Cycas in a European climate classes all the Northern Territory species except Cycas angulata as “very difficult”; it does not name Cycas armstrongii, which falls into that group by its origin. According to the author, the Northern Territory Cycas are temperamental even in the subtropical climate of the Sunshine Coast in Queensland, whereas they thrive at Nong Nooch Tropical Garden in Thailand, where it is hot almost all year round. He also points out that in Europe the coldest months are the wettest, the reverse of the climate these plants come from, and he advises protecting plants kept outdoors from excess water during winter rain. By simple deduction, the same reversal applies to other climates with winter rain, such as coastal California or southern Australia.

Container cultivation

Lavaud’s general advice for the Australian Cycas applies to a pot under glass: a substrate with a high mineral content, full sun or the brightest spot in the glasshouse, generous watering in hot weather, reduced when temperatures fall below 20 °C (68 °F) and stopped when night minimums drop below 10 °C (50 °F), because watering in cool conditions carries a very high risk of root rot. Allow for the deciduous habit described by Hill: a plant that drops its leaves or stops growing is resting, and it should not be pushed with water while it is dormant.

In cultivation in the UK, US and Australia

In short: a few botanic gardens in the United States and Australia hold the species, none in Britain on the records found. The examples below come from the gardens’ own catalogues; a garden listed in the BGCI PlantSearch database (consulted in October 2026) has declared the species at some point, which does not prove that a living plant is there today.

  • United States: the online catalogue of Montgomery Botanical Center (Coral Gables, Florida) lists one accession of Cycas armstrongii, whose number points to 1996. BGCI PlantSearch also lists it for the University of California Botanical Garden at Berkeley and, in Hawaii, for the Honolulu Botanical Gardens, the National Tropical Botanical Garden and the botanical garden of the University of Hawaii at Hilo.
  • Australia: the Royal Botanic Garden Sydney lists one plant of wild origin (accession of 2015) in its Garden Explorer (updated 1 October 2026); the Australian Botanic Garden Mount Annan lists the species in BGCI PlantSearch. No species list is documented for the George Brown Darwin Botanic Gardens, where the plants photographed in 2016 grow.
  • United Kingdom: no British garden lists the species.

No report from a grower in the United Kingdom, the United States or Australia is documented on the English-language forums that could be read.

Propagation

Like all Cycas, Cycas armstrongii is dioecious: a male and a female plant are needed to obtain fertile seed. No germination data specific to the species has been published. In the wild, plants branch from the underground caudex after fire (Hill, 1996); no propagation from offsets in cultivation is documented. Seeds are exempt from CITES under annotation #4; ask for seed collected from cultivated plants.

Pests and diseases

No pest or disease specific to the species is documented. Scale insects, including Aulacaspis yasumatsui, threaten all cultivated Cycas, but no attack on this species in particular has been reported.

Cold hardiness

There is no measured figure for the cold tolerance of Cycas armstrongii. Its whole range lies in the monsoonal tropics of northern Australia, and nothing allows us to say that it tolerates frost. No USDA zone for the species is given by any published source.

No cultivation result, in the ground or in a pot, has been published from the Mediterranean, Europe, the United States or temperate Australia, and no dated frost episode concerns it. On the French forum Fous de palmiers the species appears only in the Darwin photographs and in an identification thread, with no data on cold; the Jardin zoologique tropical in the south of France has published no observation of it. The only known cultivation records come from the tropics. Without data for the species, grow it in a heated glasshouse outside the tropics.

Toxicity and precautions

No toxicity analysis specific to Cycas armstrongii is documented, but the Flora of Australia reports that the seeds were eaten only after long and careful preparation and cooking, a sign that they are toxic raw. The risk is well documented in the genus: Cycas tissues, and the seeds above all, contain cycasin, which is converted during digestion into methylazoxymethanol, toxic to the liver and nervous system, along with a neurotoxic amino acid, BMAA. The ASPCA lists Cycas revoluta among plants poisonous to dogs, cats and horses. Treat the plant as poisonous: keep seeds and leaves out of reach of children and pets, and wear gloves to clean the seeds. If any part is swallowed, contact a poison centre, a doctor or a vet straight away.

Traditional uses

The Flora of Australia sums up the Aboriginal uses of the species as follows: seeds eaten or ground into flour after long and careful preparation and cooking; a preparation of male cones and human urine used as an antiseptic for spear wounds; seeds with strong antibiotic activity. These preparations must not be copied under any circumstances.

Frequently asked questions

Why does Cycas armstrongii lose its leaves?

Because it is facultatively deciduous: in its monsoonal habitat it often drops its leaves at the end of the dry season, and fires add to the effect. In cultivation it may keep them longer, but it still goes through a resting phase.

Where does it grow?

In Australia’s Northern Territory, around Darwin, from the Adelaide River to the Finniss River, with small populations further south and on Melville Island.

How can it be told apart from Cycas conferta?

By its shorter petioles, more often spiny, and its glossy leaflets, paler beneath and more widely spaced; Cycas conferta has crowded, dull leaflets that are alike on both surfaces.

Are Cycas armstrongii and Cycas maconochiei the same species?

Possibly. A 2025–2026 genetic study found minimal differentiation and concluded that they probably represent one variable species, but no formal revision has yet merged them.

Is it threatened?

Yes, according to the IUCN, which classes it as Vulnerable (2022), even though it is one of the most abundant cycads in the world; it is also protected under Northern Territory law and by CITES.

Can it be grown outdoors in the UK or the cooler parts of the US?

Almost certainly not. There is no data on its cold tolerance, the species is tropical, and the nurseryman Simon Lavaud places the Northern Territory Cycas among the hardest to grow in Europe; it needs a heated glasshouse in winter.

Is it poisonous?

Yes, it should be treated as such. Aboriginal people ate the seeds only after long preparation, and the Cycas that have been studied contain dangerous toxins, especially in the seeds.

Further reading on Succulentes

Sources

Reference sites

  • The World List of Cycads (version 2026.08.15), Cycas armstrongii: https://cycadlist.org/scientific_name/59 — author, protologue, type, distribution, IUCN status (VU A4cde), etymology after Haynes (2022). Checklist of the IUCN Cycad Specialist Group, reliable; consulted in October 2026.
  • Flora of Australia online (ABRS, after Hill 1998, Flora of Australia vol. 48), Cycas armstrongii: https://profiles.ala.org.au/opus/foa/profile/Cycas%20armstrongii — description, distribution, habitat, deciduous habit, statuses (IUCN and Northern Territory), Aboriginal uses. National reference flora, reliable; consulted directly.
  • Australian National Herbarium, biographical notes, “Armstrong, John”: https://www.anbg.gov.au/biography/armstrong-john.html — death at Kupang (Timor) on 21 January 1847, unknown date of birth, Australian collections from 1838 to August 1846, after Orchard (1999) and George (2009). National botanical institution, reliable; consulted on 18 October 2026.
  • PACSOA Wiki, Cycas armstrongii: https://pacsoa.org.au/wiki/index.php/Cycas_armstrongii — local name “fire-fern” for the new fronds produced after fire by this species and Cycas calcicola, from an article by L. P. Butt in Palms & Cycads no. 27 (1990). Wiki of a specialist society; used only for the common name, the rest of the page (synonymy with Cycas lane-poolei, frond count) being contradicted by Hill (1996); consulted on 19 October 2026.
  • Royal Botanic Garden Sydney, Garden Explorer, Cycas armstrongii: https://rbgsydney.gardenexplorer.org/taxon-858.aspx — one plant of wild origin, accession S2015-0739, label “Zamia palm, fire fern”. Garden’s own catalogue (updated 1 October 2026), reliable for presence; consulted on 19 October 2026.
  • Montgomery Botanical Center, living collection catalogue (BG-BASE “Montgomery – Cycads”, hosted by RBGE): https://www.montgomerybotanical.org/research/living-collection/ — one accession (96290). Garden’s own catalogue, reliable for presence; consulted on 19 October 2026.
  • BGCI PlantSearch: https://plantsearch.bgci.org/ — gardens declaring the species (UC Berkeley, Honolulu, National Tropical Botanical Garden, University of Hawaii at Hilo, Mount Annan). Global database of garden lists, undated per garden: a listing does not prove a living plant; consulted in October 2026.
  • Simon Lavaud (2021), “Quelques données sur les bases de la culture des Cycas australiens en climat européen”, Cycadales.eu: https://cycadales.eu/wp-content/uploads/2021/04/Bases-de-la-culture-des-Cycas-australiens.pdf — Northern Territory species (except Cycas angulata) classed as “very difficult”, general advice on substrate, light, watering and winter rain; does not name Cycas armstrongii. Note by a specialist nurseryman based on his own experience (in French); full text consulted.
  • Fous de palmiers forum, “[AUS] Cycas / Darwin Botanical Garden – Australie (NT)” (September 2016): https://www.fousdepalmiers.fr/html/forum/viewtopic.php?t=21346 — photographs of plants grown at the Darwin botanic garden. French collectors’ forum; background information only.
  • Species+ (UNEP-WCMC), Cycas armstrongii: https://speciesplus.net/species#/taxon_concepts/29593/legal — CITES Appendix II, EU Annex B. Official CITES and EU database, reliable; consulted on 12 October 2026.
  • Commission Regulation (EU) 2026/1383: https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=OJ:L_202601383 — annexes to Regulation (EC) No 338/97 in force since 29 June 2026. Official text.
  • ASPCA Animal Poison Control, Sago Palm: https://www.aspca.org/pet-care/aspca-poison-control/toxic-and-non-toxic-plants/sago-palm — toxicity of Cycas revoluta to dogs, cats and horses. American animal poison control centre, reliable for the genus.

Bibliography

  • Miquel, F.A.W. (1868). Archives Néerlandaises des Sciences Exactes et Naturelles 3(5): 235–236. [Protologue; not consulted directly, reference taken from Hill (1996) and The World List of Cycads.]
  • Hill, K.D. (1996). A taxonomic revision of the genus Cycas (Cycadaceae) in Australia. Telopea 7(1): 1–64. doi:10.7751/telopea19963040 [Reference revision for Australia: description, deciduous habit, hybrids, typification, history of the name; full text consulted.]
  • Hill, K.D. (1998). Cycadaceae. In: McCarthy, P.M. (ed.), Flora of Australia 48: 597–635. CSIRO Publishing, Melbourne. [Consulted in its updated online version (profiles.ala.org.au).]
  • Laubenfels, D.J. de & Adema, F. (1998). A taxonomic revision of the genera Cycas and Epicycas gen. nov. (Cycadaceae). Blumea 43(2): 351–400. [Broad concept of the species, including Cycas papuana; full text consulted.]
  • Clugston, J.A.R., Cuff, N.J., Chong, C., Calonje, M., Claravall, K., Gallagher, R.V., Henwood, M., Kenicer, G.J., Milne, R. & Ruhsam, M. (2026). Low genetic differentiation among morphologically distinct Cycas species informs the delineation of conservation management units. Annals of Botany 137(2): 415–430. doi:10.1093/aob/mcaf276 [RADseq study of Cycas armstrongii and Cycas maconochiei subsp. maconochiei, published online in November 2025; abstract consulted.]
  • Bösenberg, J.D. (2022). Cycas armstrongii. The IUCN Red List of Threatened Species 2022: e.T41987A67335340 (correction 2023: e.T41987A243413981). [Red List assessment; references checked on Crossref, category taken from The World List of Cycads and the Flora of Australia, page not consulted (access blocked).]
  • Butt, L.P. (1990). [Article on Cycas armstrongii and Cycas calcicola.] Palms & Cycads 27. [Known through its reproduction on the PACSOA wiki; used only for the local name “fire-fern”.]
  • Calonje, M., Stevenson, D.W. & Osborne, R. (2026). The World List of Cycads, version 2026.08.15. Montgomery Botanical Center, Coral Gables. doi:10.5281/zenodo.21958524 [Reference checklist; consulted directly.]