Dracaena inexpectata is an arborescent dragon tree described in 2025 by Joël Lodé, on the basis of field observations carried out on the island of Socotra (Yemen) between 2008 and 2024. It grows side by side with the famous Dracaena cinnabari, from which it departs at first glance: it lacks the flat umbrella crown that makes the Socotran dragon tree instantly recognisable, and its leaves are roughly three times longer, flexible and arching. Within the genus Dracaena, this newcomer joins the small group of arborescent dragon trees repeatedly compared with one another in the literature: Dracaena draco, Dracaena tamaranae, Dracaena cinnabari and now Dracaena inexpectata.
One caveat belongs at the top of this page. At the time of writing, the name Dracaena inexpectata appears neither in the IPNI nomenclatural index nor in POWO (Plants of the World Online, Kew), the sole nomenclatural authority followed on succulentes.net. The original description appeared in a specialist journal not indexed by the major databases, rests on type material limited to leaves, and is supported by no molecular data whatsoever. What follows therefore describes a striking and well-documented morphological entity whose taxonomic standing has yet to be settled.
How to Recognise Dracaena inexpectata ?
Habit is the first clue. Dracaena cinnabari builds a dense, flattened, upturned-umbrella crown through strictly dichotomous branching. Dracaena inexpectata instead branches from the base into elongated, unequal, frequently forked and noticeably thicker limbs, each carrying a discrete globose rosette; these rosettes never coalesce into a single parasol. The resulting outline recalls the Dracaena draco populations of north-western La Palma, in the Canary Islands.
The foliage supplies the sharpest diagnostic characters:
- Leaves 60 to 92 cm long and 17 to 25 mm wide, against 22 to 29 cm long and 12 to 15 mm wide in Dracaena cinnabari.
- Blades flexible and flattened, arching downwards, whereas Dracaena cinnabari bears rigid, thick, almost succulent and stiffly held leaves.
- A markedly broader leaf base — the “spoon” — than in Dracaena cinnabari.
- Trunk reaching 1.80 m in diameter, branching from the base upwards.
The inflorescence has never been seen in flower. Only dried spikes, still lodged in the rosette or fallen to the ground, have been examined. They appear more erect than those of Dracaena cinnabari, yet shorter and looser.
Hybrids
No hybridisation has been recorded between Dracaena inexpectata and Dracaena cinnabari, despite the two growing intermingled. The describing author reports the opposite situation: unbroken sterility. Across observation campaigns spanning February, May, June, July, October and November, not a single fruit and not a single seed was found, while neighbouring Dracaena cinnabari were fruiting heavily.
That consistent sterility deserves emphasis, because it admits more than one reading. Shifted phenology or pollen incompatibility with the neighbouring species, as the author proposes, is one possibility. The alternative — that these scattered, genetically isolated individuals are of hybrid origin, or sterile somatic variants — cannot be dismissed without DNA evidence. Settling that question was precisely what the molecular study, which failed, was meant to achieve.
Possible Confusion
- Dracaena cinnabari: the sympatric species, and the only realistic confusion in the field. Umbrella crown, even dichotomous branching, short rigid leaves, abundant fruiting.
- Dracaena draco: the Canarian dragon tree, similar in outline but absent from Socotra. The likeness to La Palma specimens is striking, and in fact prompted the description.
- Dracaena tamaranae: endemic to Gran Canaria, with a slender trunk and a much reduced crown. This was the author’s first comparison, before he settled on a Dracaena draco branching pattern instead.
- Dracaena serrulata: the dragon tree of southern Arabia and Dhofar, which likewise carries longer and more flexible leaves than Dracaena cinnabari. No comparison was drawn in the original description, and one would be worth making.
Taxonomy
Dracaena inexpectata Lodé, sp. nov., published in Cactus-Adventures International No. 124: 28-57 (2025), with the description in English on p. 30 and in French on p. 40. Family Asparagaceae, subfamily Nolinoideae.
Type: Socotra, Dixam, descent to Wadi Dirhur, 19 May 2024, J. Lodé s.n. (MGC-Cormof 93600, leaves), Málaga herbarium.
Etymology: from the Latin inexpectata, “unexpected”, referring to the wholly unforeseen nature of the discovery.
Three reservations attach to this name and should be known to the reader:
- No molecular data exist. The leaves collected in 2024 and analysed at the University of Almería produced no usable result. Any affinity with Dracaena cinnabari therefore rests on morphology and co-occurrence alone.
- The description falls short of current standards: flowers and fruits unknown, type material restricted to leaves, no precise coordinates, no altitude, no population count.
- The name is indexed by neither IPNI nor POWO as of consultation. The protologue introduces the type simply as “Type:”, without “holotypus” or an explicit equivalent, which may raise a question of valid publication under the International Code of Nomenclature. This point calls for verification with the nomenclatural bodies.
Until then, the prudent course is to treat Dracaena inexpectata as a remarkable Socotran morphotype deserving monitoring and genetic study, rather than as an established species.
In Habitat
The taxon is known from the Dixam plateau and the descent towards Wadi Dirhur, together with Wadi Ayaft, where one or more specimens were photographed by Mitsuko Shingai. It shares its ground with Dracaena cinnabari, on the same limestone plateaux and mid-altitude slopes, under a climate governed by the monsoon and by seasonal mists that offset the scarcity of rainfall.
Occurrence is described as extremely rare and sporadic: a handful of scattered individuals, genetically isolated from one another, though all in good health. The Socotra archipelago has been a UNESCO World Heritage Site since 2008 and ranks among the densest centres of plant endemism on the planet, which makes the survival of undescribed relict entities entirely plausible.
Outdoor / In-ground Cultivation
Not a single specimen of Dracaena inexpectata is in cultivation anywhere, and no seed has ever been collected. What follows is therefore inference drawn from the handling of Dracaena cinnabari, with which the taxon shares its habitat exactly.
Planted out, a Socotran dragon tree wants full exposure and a sharply draining mineral bed: lava grit, pumice and coarse sand, with organic matter kept to a minimum. A raised or sloping position pays for itself, since the enemy is standing water at the collar rather than drought. Growth is slow, especially over the first years, before the trunk begins to build. Open ground suits only frost-free, low-humidity climates; elsewhere, container growing is the realistic route.
Container Cultivation
A deep pot suits the strong initial taproot better than a wide one. The same gritty mix applies, topped with a mineral mulch that keeps the collar dry. Water generously but infrequently through the growing season, letting the substrate dry right through between applications, and reduce to almost nothing once growth slows. Under glass or in a conservatory, steady ventilation and moderate humidity come closer to conditions on the Dixam plateau than any enclosed, stagnant atmosphere. Repot infrequently — these plants resent root disturbance and recover from it slowly.
Propagation
Here the question has no practical answer. Since fruiting has never been observed, there is neither seed nor any source of seedlings. Trunk or branch cuttings, workable on several dragon trees, have never been attempted on this taxon in any documented trial.
The route proposed by the describing author is in vitro culture, already applied successfully to Dracaena cinnabari. It would require vegetative material collected under permit from the Yemeni authorities, which — given the administrative difficulties encountered over a simple leaf collection — remains the principal obstacle.
Pests and Diseases
Nothing is on record for this taxon. Arborescent dragon trees in cultivation are classically troubled by mealybugs and scale insects lodged in the leaf axils, and by spider mites in hot dry air. Collar and root rots, following an over-retentive substrate or winter watering, remain the commonest cause of loss among young plants.
Cold Hardiness
No figures exist for Dracaena inexpectata, in habitat or in cultivation, and it would be unwise to lend it by analogy a hardiness nobody has measured. On the Dixam plateau it grows in a tropical upland setting where frost does not occur. In the absence of observation, the only sensible working rule is to treat it as Dracaena cinnabari is treated: a dragon tree for mild climates, to be overwintered above freezing.
Traditional Uses
Dragon’s blood, the red resin tapped from the trunk, has been harvested since antiquity from Dracaena cinnabari for medicine, dyeing and varnish. Nothing in the published observations indicates that Dracaena inexpectata yields a comparable resin, or that islanders distinguish it from the common dragon tree at all. Its ethnobotanical standing remains an open question.
FAQ
How many individuals of Dracaena inexpectata are known? No figure has been published. The description speaks of a very rare and sporadic presence in the Dixam region and at Wadi Ayaft, without a census. Compiling an inventory is among the priorities the author sets out.
Can seeds or plants be obtained? No. No seed has ever been found, the observed plants being apparently sterile, and no material is in cultivation. Any commercial offer under this name should be treated with the greatest suspicion.
Is it genuinely a distinct species? The matter is unsettled. The morphological differences from Dracaena cinnabari are dramatic and consistent, but the absence of genetic data, flowers and fruits makes it impossible to choose between a distinct species, a sterile hybrid and a morphological variant.
What is the easiest character to check in the field? The leaf. Flexible, flattened, 60 to 92 cm long in Dracaena inexpectata; rigid, thick, 22 to 29 cm long in Dracaena cinnabari. The habit — separate globose rosettes rather than a single umbrella — confirms it at a glance.
Is the taxon threatened? No IUCN assessment exists. Given its extreme rarity, its apparent sterility, and the grazing pressure already impairing regeneration of Dracaena cinnabari on the island, the author envisages a relict situation comparable to that of Dracaena tamaranae.
Reference Sites
https://www.cactus-aventures.com https://powo.science.kew.org https://www.ipni.org
Bibliography
Lodé, J. 2025. A New Species of Dracaena in Socotra / Une nouvelle espèce de Dracaena à Socotra / Una nueva especie de Dracaena en Socotra. Cactus-Adventures International 124: 28-57. [original trilingual description; protologue p. 30 (English), p. 40 (French), p. 50 (Spanish)]
Hubálková, I. 2011. Prediction of Dragon’s Blood Tree (Dracaena cinnabari Balf.) stand sample density on Soqotra Island (key study). Journal of Landscape Ecology 4(2): 5-17.
Hubálková, I., Maděra, P. & Volařík, D. 2017. Growth dynamics of Dracaena cinnabari under controlled conditions as the most effective way to protect endangered species. Saudi Journal of Biological Sciences 24(7): 1445-1452. [key reference for ex situ propagation]
Kunkel, G. 1973. Novedades en la Flora de las Islas Orientales, con una nota sobre Dracaena draco en Gran Canaria. Cuadernos de Botánica Canaria 17: 35-37.
Maděra, P., Forrest, A., Hanáček, P., Vahalík, P., Gebauer, R., Plichta, R., Jupa, R., Van Rensburg, J.J., Morris, M., Nadezhdina, N. et al. 2020. What We Know and What We Do Not Know about Dragon Trees? Forests 11(2): 236. https://doi.org/10.3390/f11020236 [standard review of the arborescent dragon trees]
Marrero, A., Almeida, R.S. & González-Martín, M. 1998. A new species of the wild dragon tree, Dracaena (Dracaenaceae) from Gran Canaria and its taxonomic and biogeographic implications. Botanical Journal of the Linnean Society 128: 291-314. [description of Dracaena tamaranae]
Lu, P.-L. & Morden, C. 2010. Phylogenetics of the plant genera Dracaena and Pleomele (Asparagaceae). Botanica Orientalis, Journal of Plant Science 7: 64-72.
Petroncini, S., Marcheselli, M.P. & Bruschi, P. 2003. Preliminary aspects of the anatomy and genetic variability of Dracaena cinnabari, Balf. fil: a threatened endemic species of Socotra Island, South Yemen. Journal of Agriculture and Environment for International Development 97(3/4): 185-195.