The Generic Affinity of Echidnium spruceanum Schott and Its Placement in Dracontium (Araceae) Guanghua Zhu Missouri Botanical Garden, P.O. Box 299, St. Louis, Missouri 63166, U.S.A.ABSTRACT. Echidnium spruceanum Schott istransferred to Dracontium as D. spruceanum(Schott) G. Zhu. The original observation of uniloc-ular ovaries with two ovules per locule on the ho-lotype is here considered to have been erroneous.Dracontium carderi Hooker f., D. costaricense En-gler, D. trianae Engler, D. loretense K. Krause, andD. ornatum K. Krause are placed in synonymy un-der D. spruceanum. A lectotype is designated forthe name D. trianae. A year after Schott (1857) described the mono-typic genus Echidnium, based on E. schomburghii, he described a second species, E. spruceanum (Schott, 1858). He asserted that E. spruceanum dif-fers from Dracontium in having unilocular ovaries with two ovules. Based on these same characters, Engler (1878: 118) transferred this species to Cyr-tosperma. It has since remained in Cyrtosperma, al-though Engler (1889: 124) transferred the genus Echidnium to Dracontium as Dracontium sect. Echidnium Engler. The type of E. spruceanum has a three-parted, highly subdivided leaf typical of Dracontium, but not of Cyrtosperma, which has sim-ple hastate to sagittate leaves. Hay (1988: 457) re-ferred E. spruceanum to Dracontium, based on his notion that the number of ovarian locules cannot be used as a generic character in this group, but he did not make a new combination. Based on the same assumption, Bogner (1985) accepted Echid-nium in the synonymy of Dracontium and made a new combination, but the species E. spruceanum was not discussed. Spruce 2406 (K) is apparently the single element studied by Schott (1858) when he described E. spruceanum, and thus it is the holotype of the name. This specimen was annotated by Schott, although no collector and number but only the herbarium ("Herb. Hook.") were cited in the protologue. The supposedly unilocular ovary of Echidniumspruceanum may be spurious. Because of the un-derdeveloped spadix and the poor condition of thespecimen, ovary characters cannot be evaluatedfrom the holotype (Richard Keating, pers. comm.).The holotype agrees overall with the genus Dracon-tium as noted above, and there is no doubt that thisspecies should be properly placed in Dracontium.Unilocular ovaries do not occur in Dracontium andthis genus never has more than one ovule in eachlocule (Zhu, 1995). Specimens clearly conspecificwith Spruce 2406 never have unilocular ovaries nortwo or more ovules per locule. The original obser-vation of unilocular ovaries with two ovules per loc-ule on the holotype is here considered to have beenerroneous. Epitypification is not necessary becauseidentification of this species does not depend onovarian characters (Greuter et al., 1994). Dracontium spruceanum, as here circumscribed,is the most widely distributed species in the genus.It ranges from the Talamanca lowands on the At-lantic slope of Costa Rica to the Choc6 region onthe Pacific slope of Colombia, and throughout theAmazonian portions of Colombia, Ecuador, Peru,and Brazil. This species can be identifed by its longpeduncle, erect to slightly arching spathe, which isgradually acuminate above and has its marginsbroadly overlapping near the base, and by its trans-lucent area of the inner spathe surface two to fourtimes higher than the spadix (Zhu, 1995). Subsequent to Schott's (1858) original publica-tion, Dracontium spruceanum was redescribed fivetimes, as D. carderi Hooker f., D. costaricense En-gler, D. trianae Engler, D. loretense K. Krause, andD. ornatum K. Krause. These names have been fre-quently used for specimens from different locali-ties. The foregoing names are for the first time hereplaced in synonymy under Dracontium spruceanum,the nomenclature of which may be summarized asfollows:Dracontium spruceanum (Schott) G. Zhu, comb. nov. Basionym: Echidnium spruceanum Schott, Oesterr. Bot. Z. 8: 350. 1858. Cyrtosperma spruceanum (Schott) Engler in Mart., Fl. bras. III. 2: 118. 1878. TYPE: Brazil. Amazonas: Sao Gabriel, Spruce 2406 (holotype, K).Dracontium carderi Hooker f., Bot. Mag. t. 6523. 1880. Syn. nov. TYPE: Cultivated plant at Royal Botanical Gardens, Kew, originally collected by Carder in Co-NOVON 6: 308-309. 1996. _._
BioStor
