Most Critically Endangered cycads are species that were once healthy and have been driven to the edge by collectors. Encephalartos dyerianus — the Lillie cycad — is something rarer and more troubling: a species that may have been in slow demographic collapse before collectors accelerated the process. PlantZAfrica provides the key sentence: “historically this species was never abundant due to a lack of seedling regeneration which is hypothetically caused by the absence of suitable insect pollinators.” If this hypothesis is correct, then E. dyerianus was already a walking dead species — a population of aging adults producing cones that no insect visited, seeds that were never fertilised, seedlings that never appeared — long before the first collector climbed the granite koppie near Mica and began digging plants out of the ground.
The collectors came anyway. In the 1970s, the population was estimated at approximately 600 individuals. In January 2008, 107 plants were removed in a single poaching event — one sixth of the entire species, taken in a matter of days. The SANBI Red List notes that “poaching incidents on the unprotected hill” have continued since. What remains is “one viable subpopulation and a few isolated individuals.” Encephalartos dyerianus is the largest, most imposing member of the eugene-maraisii complex of blue-leaved Encephalartos: trunks reaching 4–6 m, dense crowns of long silvery-blue fronds, and striking blue-green cones. It is also a species being destroyed from two directions at once — from above, by collectors who take the adults; and from below, by a reproductive failure that prevents replacements from growing.
Taxonomy and nomenclature
Encephalartos dyerianus Lavranos & D.L. Goode was first published in 1988 in the Bulletin du Jardin Botanique National de Belgique (volume 58(1–2): 219–224) — the same volume that carried the original description of Encephalartos dolomiticus. The holotype (J. Smit s.n., PRE 32864) was collected from the farm Lilie (Lillie), north of Mica, in the lowveld, on a single granite hill, at approximately 700 m elevation, in February 1966 — twenty-two years before the species was formally described. The holotype is deposited at PRE (Pretoria).
The epithet dyerianus honours Dr Robert Allen Dyer (1900–1987), one of the foundational figures in South African cycad taxonomy. Dyer served as director of the Botanical Research Institute in Pretoria (now SANBI) from 1944 to 1963, started the Pretoria National Botanical Garden, and described several Encephalartos species — including Encephalartos heenanii. His monograph of South African cycads (Bothalia 8: 405–515, 1965) remains a reference work. Naming a species after Dyer was a tribute to the man who had laid much of the groundwork for modern Encephalartos systematics. It is an irony that the species bearing his name — a man who dedicated his career to documenting South African cycads — is now among the most threatened of all the species he worked to describe.
A taxonomic synonym exists: Encephalartos graniticola Vorster, Robbertse & S. van der Westh. (South African Journal of Botany 54(4): 363–366, 1988) — published the same year as Lavranos & Goode’s description. The epithet graniticola (“granite-dweller”) described the species’ habitat with admirable precision — it grows exclusively on granite outcrops — but lost out on priority to dyerianus. Tree SA records the earlier informal name as Encephalartos graniticolus (with the masculine Latin ending).
The species belongs to the eugene-maraisii complex but occupies a position distinct from the other members. It grows at lower altitude (~700 m) than any other species in the complex: compare eugene-maraisii at 1400–1500 m, dolomiticus at 1100–1500 m, middelburgensis at 1400–1600 m. Its climate is hotter and drier — lowveld, not Highveld. And it is the tallest species in the complex by a substantial margin (4–6 m trunk vs. 2.5–4 m for eugene-maraisii, 0.8–2 m for dolomiticus). PlantZAfrica notes that dyerianus “is closely related to and often confused with E. dolomiticus, E. eugene-maraisii, E. middelburgensis and E. cupidus.”
Common names: Lillie cycad, Lillie’s cycad (English, after the farm Lilie/Lillie where the type was collected); lowveld cycad (English); Lillie-broodboom, Lillie-se-broodboom (Afrikaans).
Morphological description
Habit and caudex: Encephalartos dyerianus is the largest member of the eugene-maraisii complex — a substantial tree cycad. The trunk is erect, reaching 4–5 m in height (PlantZAfrica gives up to 5 m × 500 mm diameter; Exclusive Cycads SA reports stems up to 6 m; Tree SA gives 4.5 m × 60 cm). Long stems tend to lean. The trunk is unbranched above ground but produces suckers from the base, forming clumps. Much of the stem is covered by a uniform layer of persistent leaf bases — the geometric pattern typical of arborescent Encephalartos.
Leaves: The fronds are 130–170 cm long (PlantZAfrica gives up to 1.3 m; Africa Cycads gives 140–170 cm), pale bluish-green to blue-silver with a matte finish. The crown is notably very dense — PlantZAfrica emphasises “a very dense leaf canopy,” and this extraordinary density is one of the species’ most striking characters. No other species in the eugene-maraisii complex produces as many simultaneous leaves in as tight a crown. The effect is of a solid hemisphere of blue-silver foliage sitting atop the trunk — a visual impact that justifies the plant’s extreme desirability among collectors.
PlantZAfrica provides a precise description of the leaf curvature: the leaves are “curved outwards for a short distance just above the leaf base and then curving gently inwards along the remainder of the leaf.” This double curve — outward near the base, then inward along the length — creates a graceful vase-shaped silhouette when viewed from the side, distinct from the straight leaves of middelburgensis and the twisted leaves of dolomiticus.
The median leaflets are 17–24 cm long and 13–18 mm wide (Africa Cycads), lanceolate, with toothed edges and an insertion angle of 45–80° on the rachis — creating a keeled leaf cross-section. Basal leaflets are reduced to thorns (up to 6 spines on the petiole). The rachis is normally straight and stiff but may be slightly twisted. Tree SA notes that the cone scales are “covered with protuberances” — a verrucose character shared with dolomiticus but not with the smooth-coned nubimontanus.
Reproductive structures: Male cones are ovoid, blue-green or yellow, 30–50 cm long and 9–12 cm wide — large for the complex. Female cones are ovoid, blue-green or yellow, 30–60 cm long and 10–20 cm wide. Both sexes’ cones change from blue-green to yellowish-green at maturity (Exclusive Cycads SA). Seeds are oval, approximately 4.2 × 3 cm, with a scarlet sarcotesta — bright red seeds that attract birds, monkeys, and baboons for dispersal.
Distribution and natural habitat
Encephalartos dyerianus is endemic to the eastern lowveld of Limpopo Province, South Africa. The species is restricted to one or two granite outcrops (koppies) near the settlements of Mica and Gravelotte. The type locality is the farm Lilie (Lillie), north of Mica. Exclusive Cycads SA records “two granite outcrops in the eastern part of Limpopo.” Tree SA places some populations within the boundaries of the Kruger National Park in Mpumalanga — a significant conservation detail, as plants within the Park receive formal protection.
The altitude is approximately 600–700 m — substantially lower than the 1000–1500+ m habitats of the other blue-leaved species in the complex. The vegetation association is Gravelotte Rocky Bushveld (Wikipedia), open grassland and shrubland on the slopes of low granite koppies. The species may coexist with mopane trees (Colophospermum mopane) — a typical lowveld association that no other species in the eugene-maraisii complex shares. The rainfall is approximately 500 mm per year (PlantZAfrica), making this the driest habitat of any species in the complex.
The granite substrate is significant. Granite is an igneous rock — silica-rich, producing acidic, sandy, well-drained soils when it weathers. This is the opposite of the dolomite (alkaline, calcium-magnesium-rich) substrate on which dolomiticus grows. The two species’ names — graniticola (“granite-dweller”) for the synonym, dolomiticus (“of dolomite”) for the sister species — encode the fundamental geochemical distinction between their habitats.
The pollinator hypothesis — why the species was already dying
PlantZAfrica’s statement deserves to be quoted in full context: “historically this species was never abundant due to a lack of seedling regeneration which is hypothetically caused by the absence of suitable insect pollinators.”
This hypothesis — if correct — has profound implications. All Encephalartos species are insect-pollinated. The pollination system is highly specific: most species are pollinated by one or a few species of beetle (typically weevils of the genus Antliarhinus or cucujoid beetles), which are attracted to the cones by thermogenic heating and volatile organic compounds. The beetles feed on cone tissue and pollen, and in doing so transfer pollen between male and female cones. Without the specific beetle pollinator, the cones produce unfertilised ovules, and no viable seed is set.
If dyerianus has lost its pollinator — or never had one at this particular site — then the species has been reproductively sterile in the wild for an unknown period. The existing population consists entirely of aging adults that cannot replace themselves. Even without a single collector ever touching the population, it was doomed to eventual extinction as the oldest plants died without producing seedlings. The collectors simply accelerated a process that was already underway — they removed the adults faster than the adults would have died naturally, but the endpoint was the same.
The hypothesis raises a further question: why would the pollinator be absent? Several possibilities exist. The pollinator species may have gone locally extinct due to habitat changes (fire regime alteration, invasion by alien plants, pesticide drift from nearby farmland). Alternatively, the dyerianus population may be a relict — a fragment of a formerly larger population that once supported a pollinator community, but which has contracted to below the minimum population size needed to sustain its beetle mutualists. Or the population at the Lillie hill may be a founder colony established by long-distance seed dispersal, at a site beyond the range of the pollinator species.
Whatever the cause, the practical consequence is that hand pollination is essential for any conservation propagation effort. Wild reproduction is not occurring, and cannot occur until the pollinator question is resolved — either by identifying and reintroducing the missing beetle, or by establishing a permanent hand-pollination programme at the remaining sites.
Conservation — the ongoing haemorrhage
Encephalartos dyerianus is assessed as Critically Endangered (CR) on the IUCN Red List, under criteria B1ab(v)+2ab(v).
1970s: Population estimated at approximately 600 plants.
1980s–2000s: Severe, sustained reduction by collectors. PlantZAfrica: “the population has been severely reduced due to illegal removal of mature plants for ornamental or horticultural purposes.”
January 2008: 107 plants removed in a single poaching event. This is one of the most devastating single-incident losses documented for any Encephalartos species — 18% of the estimated 1970s population taken in one month. The SANBI Red List specifies that this occurred “on the unprotected hill” — implying that at least one of the granite koppies lacks formal conservation status.
Post-2008: “One viable subpopulation and a few isolated individuals remain” (PlantZAfrica). The SANBI Red List adds that “poaching incidents on the unprotected hill” have continued, and that it is “thought likely that it has continued to decline.”
The combination of adult poaching and the absence of natural seedling recruitment creates a population in terminal demographic decline: adults are being removed faster than they can be replaced, and even without removal, no replacement is occurring because no seeds are germinating. The population trajectory is a one-way line toward zero.
Cold hardiness
The lowveld habitat at 600–700 m is warmer than the montane habitats of the other blue-leaved species in the complex. Frost is possible but less frequent and less severe than at the 1400–1500 m altitudes of eugene-maraisii and middelburgensis.
Practical cold hardiness estimate: USDA Zone 9b–10a (−1 to −4 °C). The species is the least frost-hardy member of the eugene-maraisii complex, reflecting its lowveld origin. Light frost is tolerated; sustained freezing is not.
Caveat: Cold-hardiness reports from zones colder than 9b should be treated with caution. Young plants with subterranean or barely emergent caudices benefit from soil thermal inertia. A mature plant with a 4–6 m aerial trunk exposes its apical meristem to ambient air temperature with no soil protection — and a meristem at 5 m height receives no benefit from ground-level insulation. A single isolated success in a cool garden does not prove the species can survive identical conditions everywhere.
Cultivation — easy to grow, almost impossible to buy
Difficulty: 2/5. Africa Cycads: “a handsome, undemanding and easily grown cycad.” The species is adaptable, fast-growing (by cycad standards), and tolerant of a range of conditions. The difficulty — as with dolomiticus — is not in growing the plant but in obtaining it legally.
Light: Full sun. Africa Cycads: “In coastal and inland areas, this species prefers to be in full sun, but can tolerate some shade. For desert areas, this species should be placed in partial sun or filtered light.” The granite-koppie habitat was fully exposed.
Soil: Well-drained, granite-derived, slightly acidic. The granite substrate produces sandy, mineral-rich, well-drained soils. In cultivation, a standard free-draining cycad mix of coarse sand, loam, and pumice works well. Africa Cycads notes the species “does not do well with excessive moisture” — a critical point, as the 500 mm lowveld rainfall is drier than the habitats of most other blue species.
Watering: Moderate during the growing season, reduced during winter. The 500 mm lowveld rainfall regime is the driest in the complex. The species is drought-adapted but responds to regular watering with improved growth and denser foliage — the thick leaf canopy that makes the species so desirable requires adequate nutrition and moisture to develop fully.
Growth rate: Moderate to fast — the largest in the complex, and it reaches its size in a reasonable timeframe. Trees SA notes the species is “slow growing” but this is by tree standards; by cycad standards, dyerianus is relatively vigorous.
Landscape use: Outstanding. The combination of tall trunk (to 4–6 m), dense silvery-blue canopy, and blue-green cones makes Encephalartos dyerianus one of the most visually impressive blue cycads available for subtropical gardens. The dense leaf canopy — denser than any other species in the complex — provides year-round ornamental value. The double-curved leaf posture (outward then inward) creates a uniquely elegant crown profile.
Propagation: From seed (hand pollination mandatory — the pollinator-absence issue applies in cultivation as much as in the wild) or suckers. Standard protocols apply: collect pollen when shedding, store at −15 °C, monitor female cone-scale opening (3 days to 2 weeks), pollinate by syringe or brush. Seed cleaned, float-tested, sown on heated sand at 24–28 °C.
Comparison with the eugene-maraisii complex
| Character | E. dyerianus | E. dolomiticus | E. eugene-maraisii | E. middelburgensis |
|---|---|---|---|---|
| Altitude | 600–700 m (lowest) | 1100–1500 m | 1400–1500 m | 1400–1600 m |
| Rainfall | ~500 mm (driest) | 650–800 mm | ~700 mm | ~650 mm |
| Substrate | Granite (acidic) | Dolomite (alkaline) | Sandstone (acidic) | Various |
| Trunk | 4–6 m (tallest) | 0.8–2 m | 2.5–4 m | 1.5–3 m |
| Leaf length | 130–170 cm (longest) | 60–80 cm | 70–150 cm | 80–150 cm |
| Crown density | Very dense (diagnostic) | Full but twisted | Moderate | Moderate |
| Leaf curvature | Outward then inward (vase shape) | Spirally twisted (corkscrew) | Slightly upcurved tip | Straight, powdery bloom |
| Cold hardiness | Zone 9b–10a (least hardy) | Zone 9a–9b | Zone 8b–9a | Zone 9a |
| Pollinator status | Possibly absent (reproductive failure) | Not documented | Specific beetle (A. verdoornae) | Not documented |
| Vegetation association | Mopane woodland (unique in complex) | Open grassland on dolomite | Sandstone grassland | Mixed grassland |
| IUCN status | CR (600→declining; 107 poached 2008) | CR (possibly EW; 4 in 2019) | EN (400–620) | CR |
The Lillie hill — anatomy of a collapse
The farm Lilie (Lillie), north of the tiny settlement of Mica in the eastern lowveld of Limpopo, is an unremarkable granite koppie — one of thousands of similar inselbergs scattered across the bushveld of southern Africa. What made it unique was its cycad population: hundreds of tall, blue-silver Encephalartos dyerianus growing among the granite boulders, their dense canopies catching the lowveld light. That population has been systematically dismantled over half a century.
The SANBI Red List’s reference to “the unprotected hill” is telling. At least one of the granite outcrops where dyerianus grows has no formal conservation status — no nature reserve, no fence, no rangers. The 107 plants taken in January 2008 were removed from what was essentially an unguarded natural area. The event was documented only because someone counted the holes left behind. How many smaller poaching events went unrecorded before 2008 — a plant here, three plants there — is unknown.
The species now survives principally in collections. Africa Cycads describes it as “handsome, undemanding and easily grown” — and therein lies both the hope and the tragedy. The species is easy to propagate, easy to cultivate, and spectacularly beautiful. If sufficient legitimately propagated material exists in collections, the species can be maintained indefinitely ex situ. But the wild population — the population that evolved on those specific granite koppies, in that specific lowveld climate, with whatever remains of its ecological interactions — may be beyond recovery. The pollinators may be gone. The adults are vanishing. And the seedlings were never there to begin with.
Encephalartos dyerianus may be the next South African cycad to cross the line from Critically Endangered to Extinct in the Wild. If it does, it will be the species that was dying before it was stolen — a plant whose fate was sealed by two failures: the failure of its ecosystem to provide the pollinator it needed, and the failure of its country to protect the adults it had.
Authority websites
POWO — Plants of the World Online: https://powo.science.kew.org/…
IUCN Red List: https://www.iucnredlist.org/species/41885/121558814
World List of Cycads: https://cycadlist.org
Bibliography
Lavranos, J.J. & Goode, D.L. (1988). Encephalartos dyerianus. Bulletin du Jardin Botanique National de Belgique 58(1–2): 219–224. [Original description]
Vorster, P., Robbertse, P.J. & van der Westhuizen, S. (1988). Encephalartos graniticola. South African Journal of Botany 54(4): 363–366. [Synonym]
Dyer, R.A. (1965). The cycads of southern Africa. Bothalia 8: 405–515. [Foundational monograph by the honoree]
Grobbelaar, N. (2002). Cycads — with Special Reference to the Southern African Species. Privately published, Pretoria.
Whitelock, L.M. (2002). The Cycads. Timber Press, Portland. 374 pp.
Donaldson, J.S. (ed.) (2003). Cycads: Status Survey and Conservation Action Plan. IUCN/SSC Cycad Specialist Group, IUCN, Gland.
Donaldson, J.S. (2009). Encephalartos dyerianus. National Assessment: Red List of South African Plants.
Government Gazette (2016). Biodiversity Management Plan for 11 Critically Endangered and 4 Endangered Encephalartos species. Government Gazette 40793.
Haynes, J.L. (2022). Etymological compendium of cycad names. Phytotaxa 550(1): 1–31.