Solitary or colony-forming evergreen herbs, with subterranean erect (orthotropic) rhizomes. LEAVES 2-6, petioles smooth, unarmed, marbled at base; blades ovoid to sagittate, held erect, entire to strongly perforate. INFLORESCENCES terminal, on a peduncle shorter than petiole that sometimes is mostly underground; spathe chartaceous to subcoriaceous, forming a basal tube with a fornicate lamina, deciduous soon after anthesis; spadix much shorter than spathe, stipe very short to absent. Flowers hermaphroditic throughout the length of the spadix, tepalate, maturing in basipetal sequence, perianth and androecium tetramerous; stamens with linear filaments and extrorse bilocular anthers, dehiscence by a slit-like pore; pollen monosulcate, ellipsoid-oblong, exine subreticulate; gynoecium longer than tepals, ovate--conoid, ovary 1-2locular, locules 1-3 ovuled, placentation basal, style conoid, stigma small and not visibly nectariferous. INFRUCTESCENCE usually pendant, spathe marcescent; berries broadly obovoid, somewhat furrowed, 1-5 seeded, green to dark purple; seed campylotropous, albuminous, testa thick, verrucose to channeled. 2n = 26 (only counted for D. desciscens)
HABIT : unarmed herbs, sometimes robust (to 2m), rhizome subterranean, erect, sparsely branched, sometimes bearing a few subterranean tubercles. LEAVES : several, ± erect. PETIOLE : long, smooth to roughened-verruculate, often transversely variegated, geniculate at apex, sheath less than half petiole length. BLADE : deeply sagittate to subtripartite, often with a few perforations of irregular size between primary lateral veins; basal ribs very well-developed, primary lateral veins mostly arising near petiole insertion, very long-arcuate towards apex of each division, running into margin, higher order venation reticulate. INFLORESCENCE : solitary, appearing with leaves. PEDUNCLE : shorter than petiole. SPATHE : marcescent, tube with convolute margins, longitudinally white-striped, blade strongly fornicate, usually obscuring mouth of tube, brown-purple. SPADIX : sessile to shortly stipitate, cylindric, obtuse, shorter than spathe tube, flowering sequence basipetal. FLOWERS : bisexual, perigoniate; tepals 4, fornicate, subtruncate. STAMENS : 4, strongly exserted from flower at anthesis, anthers longer than connective, connective slender, thecae ovate-elliptic, dehiscing by pore-like apical slit. POLLEN : monosulcate, ellipsoid-oblong, small (mean 23 µm.), exine subreticulate. GYNOECIUM : ovate-conoid, ovary 2-locular, ovules 1 per locule, anatropous, placenta axile, style attenuate, longer than tepals, stigma small, button-like. BERRY : obovoid, somewhat furrowed, 1-2-seeded, dark purplish-red. SEED : reniform, attenuate towards micropyle, testa strongly dentate-cristate, thick, hard, brown, embryo curved, endosperm present.
Helophytes with deep, erect, subterranean rhizome; petiole smooth to verruculate, geniculate at apex; leaf blade deeply sagittate to subtripartite; spadix hidden by spathe; flowers bisexual, perigoniate; seed reniform, strongly dentate-cristate. Differs from Urospatha in having a fornicate (arched forward and down) spathe, leaves often sparsely perforated, and endemism to Brazil (coastal Bahia and Espirito Santo).
All collections of Dracontioides are associated with sandy soils (either flooded or not), always close to sea level. Despite observing a few populations of D. desciscens in half shade, heliophytism is the rule for both species. Robust colonies of D. desciscens were seen growing in full sunlight, as were all fertile D. salvianii. Pollinators are unknown but there is some evidence of visitation by large blowflies in inflorescences of D. desciscens in North America (Boos, 1997). Inflorescences have a smell of decaying meat, so flies could be the natural pollinators of this species. The spathe is marcescent after flowering and the seeds are involved with a gelatinous pulp. Vegetative reproduction seems to perform an important role, at least in D. desciscens. Although Boos (1997) doubted the existence of small tubercles attached in the rhizome of D. desciscens, they are very numerous in both populations I could find in Bahia and also occur in D. salvianii where they are proportionally smaller.
Dracontioides was erected by Adolph Engler in 1911 to accommodate a material from Itaparica Region (Northwestern Bahia State), originally described by Schott in 1859 as Urospatba desciscens. A second gathering was made by the Archduke Maximilian in a place named "Tijuca". A fabulous plate of this collection was prepared from living plants and published by Peyritsch (1879). This plate was reproduced in ink by Engler when he decided to transfer it to Dracontioides, based mainly on placentation type and spathe form, which is lanceolate and apically twisted in Urospatha and apically fornicate in Dracontioides.
The first important recollection of this species was done by Simon Mayo in 1977, who published the first pictures of living plants (Mayo, 1978). The subtribal position of Dracontioides was evaluated by Hay (1992), who was the first to assign the genus to the same subtribe as Anaphyllopsis and Dracontium (Dracontiineae).
After receiving a few plants of D. desciscens from Espirito Santo state, Julius Boos (1997) published the first horticultural remarks about Dracontioides, describing in rich detail its behavior in cultivation. In the same year, Mayo and collaborators (1997) presented a generic level evaluation of Dracontioides, placing it in the subfamily Lasioideae but not adopting the tribal and subtribal organization proposed by Hay.
Hay (1992) pointed out that Dracontioides should be considered to be close to Dracontium and Anaphyllopsis, two exclusively Neotropical genera. The main reason for this grouping is the occasional presence of leaf fenestrations, only occurring in these three genera in Lasioideae (although it is common in the not so closely related Monsteroideae). The presence of leaf fenestration is known in Anaphyllopsis cururuana (Hay, 1988; Goncalves, in this volume), Dracontioides (only D. desciscens) and many species of Dracontium (Hay, 1992; Zhu & Croat, 2004). The other features considered are not so consistent. Mesophylls as cataphylls was not found in Dracontioides (pers. observ., Boos, 1997) and seeds of the Dracontiineae are not consistently different from some Lasiineae, such as some Cyrtosperma species. The close relationship between Dracontioides and Dracontium was also supported by French et al. (1995) using cpDNA phylogeny. Both genera appear as sister groups, but Anaphyllopsis did not cluster in the same clade. Since this phylogeny was reconstructed using only analysis of restriction fragments, it is possible that gene sequencing can improve the resolution of the tree.