Alocasia heterophylla 'Dragon's Breath'

Where it comes from

Alocasia heterophylla (C.Presl) Merr. is accepted by Kew, published in the Philippine Journal of Science in 1908.¹² The range is not merely “the Philippines”: Kew gives it as endemic to Luzon, Mindanao and Polillo, in lowland rain forest — dipterocarp forest — to about 300 m.³ That is a real habitat with a real altitude, which turns out to be unusual among the plants on this site’s Alocasia shelf, and it is the most useful thing on this page: warm, humid, shaded, low.

Four synonyms, and they divide in a way worth getting right. Caladium heterophyllum C.Presl is the original description, and Colocasia heterophylla (C.Presl) Kunth is that same name moved to another genus — the two are homotypic, resting on one specimen. Only the other two, Alocasia manilensis and Alocasia warburgii, are separate descriptions of the plant by somebody meeting it fresh, and both are Engler’s.¹⁴⁵ So the species was independently described twice, not four times. The basionym’s date is itself unsettled: Kew prints 1828 and IPNI 1827, from the same work.⁶

‘Dragon’s Breath’ is a trade name, and an undocumented one. The International Aroid Society is the registration authority for Araceae⁷ and has no record of it; neither does the RHS.⁸ One specialist reference attributes its introduction to a named nursery,⁹ but that nursery’s own listing claims nothing of the kind, and the same reference’s page for the species does not include ‘Dragon’s Breath’ among the cultivars it recognises.¹⁰¹¹ Treat it as a selling name attached to a real species.

Naming and description

heterophylla means “with varying leaves”, and the variation is real — but not in the way this page used to claim. What is documented is that the blade shape varies, and that a single adult plant can carry both peltate and non-peltate leaves at once.¹² Lobing and texture are not recorded as varying, and the tempting story that the synonyms exist because the species is variable is not what the monographer says. Hay’s actual reasons are duller and better: A. manilensis was described from a type consisting only of seedling leaves, and A. warburgii’s Berlin types no longer survive, so he follows Engler and Merrill’s joint opinion.¹³

The leaf is a long narrow spear — narrowly hasto-sagittate, to about 27 cm, on a plant of roughly 40 cm — with a silvery or blue-green metallic wash and a fine line running just inside the margin.³¹² That line is the species’ own submarginal vein, about 3 mm in, and with three or four primary lateral veins each side it is the most reliable check on a label. The stem is decumbent to creeping — a rhizome³ — which is why the frontmatter here says rhizomatous rather than cormous; the things the hobby calls corms are cormels borne off it.

Worth knowing before shopping: this is not one of the ‘Dragon Scale’ plants, and the usual way of saying so is backwards. The leaves are not superficially similar. Alocasia baginda is broadly ovate to sub-orbicular and peltate, with pale grey bullate patches; this one is a narrow spear and explicitly not bullate.¹⁴ It is the names that are similar. One UK retailer’s listing for ‘Dragon’s Breath’ describes it as “deeply bullate” with a scale-like pattern, which is a description of the other plant.¹⁵

Fun facts

• Engler described it twice in the same paper, two pages apart. Alocasia manilensis and Alocasia warburgii both appear in Botanische Jahrbücher 25, on pages 23 and 25, published the same day in March 1898.⁴⁵ Not a case of a species being rediscovered across decades by different botanists — one man, one sitting, two names.
• “Mine doesn’t look like the pictures” cuts both ways here. The within-plant variation is genuine, so an odd leaf is often fine. But this name has a documented history of being hung on the wrong plant: Hay records it applied by various workers to distinctly different species in Java and Borneo,¹³ and the confusion worsened after two further Philippine species were separated out in 1999.¹² Check the submarginal vein before concluding your plant is merely characterful.
• Resting is a response, not a season. Kew describes the genus as seasonally dormant to evergreen,¹⁶ and the RHS frames the rest as something that happens in cooler, drier conditions.¹⁷ So leaf drop in a cold dark winter is the plant responding rather than a calendar event you must engineer — the Alocasia defaults cover what to do, which is mostly to stop watering and wait.
• It is toxic, and the sap is the part you will actually meet. All parts carry calcium oxalate; NC State rates the severity medium, and the RHS flags the sap as a skin and eye irritant.¹⁸¹⁷ Gloves when cutting, which you will be doing if you divide it.