Aliso: A Journal of Systematic and Floristic Botany Aliso: A Journal of Systematic and Floristic Botany A Revision of the Euphorbia Dioscoreoides Complex A Revision of the Euphorbia Dioscoreoides Complex

A revision of the Euphorbia dioscoreoides complex (subgenus Agaloma) is provided. Euphorbia dioscoreoides ssp. attenuata and E. eglandulosa, both from Mexico, are proposed as new; E. digitata is reduced to synonymy under E. subpeltata. Representative specimens are cited, and distributional and ecological data are provided. Se presenta una revision taxonomica del Euphorbia dioscoreoides complejo (subgenero Agaloma). Se describen una especie nueva, Euphorbia eglandulosa, y una subespecie nueva, Euphorbia dios coreoides ssp. attenuata, los dos de Mexico. Se reduce E. digitata como sinonimo de E. subpeltata. Especfmenes son citados, y datos ecologicos y distribucionales son proporcionados. Palabras clave: Euphorbia, subgenero Agaloma, Euphorbiaceae, Mexico, revision.

Euphorbia L., sensu lata, with as many as 2,000 species, is the largest genus in the family Euphorbiaceae and one of the largest among the flowering plants.The subgenus Agaloma (Raf.)House is one of nine subgenera recognized by Wheeler (1943) and is distinguished from other members of Euphorbia by its involucral appendages and minute, usually glanduliform stipules.This subgenus comprises about 150 New World species and is most diverse in Mexico and Central America where approximately 85 mostly endemic species occur.The taxa treated in this paper belong to Agaloma and include Euphorbia dioscoreoides Boiss.and its close relatives.
These taxa share the following diagnostic characters, the combination of which distinguishes them from other members of the subgenus Agaloma: leaves with long, slender, peltately attached petioles; cyathia subtended by a pair of linear-filiform opposite leaves; and invo1ucral appendages divided into three to nine segments.The leaf surfaces are essentially glabrous, and with the exception of E. dioscoreoides ssp.dioscoreoides, the proximal margin of the leaf is usually ciliate.The exact phylogenetic relationships between the individual members of the Euphorbia dioscoreoides complex are not yet elucidated, and its relationship to other members of Agaloma is also not obvious, but the long-petiolate leaves, relative lack of bracteoles be-tween the staminate flowers, and deeply pitted seeds warrant its placement in section Cyttarospermum Boiss.[Note: Many authors incorrectly apply the name Adenopetalum (Klotzsch & Garcke) Benth.& Hook.f.(published 1880) to this section.However, at the sectional rank Cyttarospermum (published 1862) has nearly twenty years priority.The type of both sections is the same, namely Euphorbia graminea Jacq.] The Euphorbia dioscoreoides complex is restricted to Mexico, ranging from Sonora to Tamaulipas south to Chiapas.As here defined it consists of three species, one comprising two subspecies.Of these, one species has gone virtually unrecognized since it was described by Sereno Watson over one hundred years ago; one species and one subspecies are here described as new.These taxa are very similar vegetatively, but can readily be separated from each other on the basis of involucral-appendage characters (Fig. 1-4).Although nearly all of the specimens that I have examined were previously identified as Euphorbia dioscoreoides, the nomenclaturally typical taxon, E. dioscoreoides ssp.dioscoreoides, is known only from a relatively small area in the states of Mexico and Michoacan.The following key will distinguish the members of this complex: Erect, leafy, taprooted annual to 1.3 m, usually 80 em or less; branchlets slender, glandular-pilose with multicellular trichomes 0.3-1.0mm long; cauline leaves alternate; stipules inconspicuous, gland-like, 0.1-0.3mm long, quickly deciduous; petioles mostly longer than the blade, slender, 1.9-7.5 em long, glandular-pilose proximally, glabrous distally, peltately attached; blades entire, membranous, penninerved, lanceolate to ovate to deltoid, 1.8-4.4em long, 1.0-3.9em wide, acute to mucronulate at the apices, rounded to truncate at the bases, the surfaces essentially glabrous and the margins glabrous or ciliate proximally with multicellular trichomes 0.2-0.4mm long; cyathia borne in axillary monochasial cymes 1.3-6.1 em long, axes of the cymes glabrous or distally glandular-pilose; subcyathialleaves opposite, linear filiform, 0.3-1.4em long, stipules inconspicuous, ca.0.2 mm long; peduncle 1-3 mm long, glabrous; involucre obconic-campanulate, 1.0-1.4mm long excluding the appendages, attenuate to truncate at base; involucra!lobes inconspicuous, ca.0.3 mm long, fimbriate at the apex; involucra!appendages 3-parted, white to green and frequently suffused with wine red coloration; involucra!glands 5, reniform, often wrinkled, the radial axis (length) 0.1-0.2mm, the tangential axis (width) 0.3-0.5 mm; bracteoles few per cyathium, filiform and often divided above; staminate flowers ca.20-30, androphores glabrous; gynophore glabrous or pubescent at its base, elongating to 7.1 mm in fruit; ovary glabrous, roundly 3-lobed; styles 3, biparted, filiform, 0.6-1.1 mm long; capsule strongly 3-lobed, 2.7-3.4 mm long, 3.1-4.0mm wide; columella 1.9-2.9mm long; seeds ecarunculate, ovoid, flattened at the base, 2.0-2.5 mm long, 1.4-1.7 mm in diameter, blackish, with numerous coarse tubercles interspersed with several regular to irregular longitudinal rows of 3-5 shallow isodiametric depressions, the sides of which are beset with numerous minute tubercles and the bottoms of which commonly contain a minute, sharply punctiform pit.Euphorbia dioscoreoides is rather widely distributed along the Pacific slope of Mexico, ranging from Sonora and Chihuahua southward to Michoacan and Mexico (Fig. 5).As here treated, it consists of two geographically segregated subspecies.These are morphologically distinct, and I have seen no intermediates.Both are summer-fall annuals whose growth corresponds with the summer monsoon season.EUPHORBIA DIOSCOREOIDES Boiss.ssp.DIOSCOREOIDES-TYPE: "Nova Hispania" (holotype G, not seen; fragment [of the holotype] F!). Leaf margins glabrous; divisions of the involucra!appendages 3, 1.4-2.4mm long, oblong to obovate, obtuse to truncate and frequently irregularly crenate at apex.An excellent illustration of this subspecies can be found in Boissier (1866: table 37).
Tropical deciduous forest and oak woodland in eastem Michoacan and the southwest portion of the state of Mexico.It is replaced in western Michoacan by ssp.attenuata.In addition to occupying different geographical ranges, subspecies attenuata and dioscoreoides also have different elevational preferences; subspecies dioscoreoides ranges from 950 m to 1550 m while s: p. attenuata primarily ranges from 90 m to 920 m.July to October.Euphorbia dioscoreoides and E. peltata are apparently based on the same Sesse and Mo~ifio collection made over two h~ndred years ago.The collectors and exact type locality of E. dioscoreoides were most likely unknown to Boissier, who described it on the basis of a specimen in Pavon's herbarium from "Nova Hispania."This specimen, the holotype, was incorporated into Boissier's herbarium, and according to Dr. Rogers McVaugh (pers. comm., 1994), it bears a printed label "Nueva Espana Herb.Pavon" and a name added by Boissier, "E.peltata."Although Sesse and Mo~ifio took part in their Mexican expedition during the late 18th and early 19th centuries, it was not until the late 19th century that their manuscripts on the flora of Mexico were published.In the meantime, their herbarium collections had been disseminated to various people and institutions.Many of the specimens are known to have found their way into Pavon's herbarium (McVaugh, 1961: 173), and, as discussed by McVaugh (1987: 169), it was Pavon's custom to include with the specimen a small ticket containing virtually no information except the name originally given to the plant by Sesse and Mo~ifio.It seems probable that Boissier added the name from the ticket onto the label, and it is therefore a fair assumption that this is the Euphorbia peltata of Sesse and Mo~ifio.A comparison between a fragment of the holotype of E. dioscoreoides and a fragment of a Sesse and Mo~ifio collection labeled as E. peltata (both at F) shows them to be identical in all respects.

EUPHORBIA
Tropical forest, oak woodland, and pine-oak woodland, commonly in disturbed habitats and frequently on limestone substrates.This species ranges from central Coahuila to southwestern Tamaulipas, southward through San Luis Potosf to Guanajuato, Queretero, and Hidalgo; it also occurs in Morelos, Guerrero, and Oaxaca (Fig. 6).Reproductive from April to November and in February, so far as is known.250-1600 m.A comparison between the type material of Euphorbia digitata and Euphorbia subpeltata reveals no differences of taxonomic merit between them, and the two are here considered synonymous.The leaves of Pringle 3272 (the type of E. subpeltata) are nearly orbicular and obtuse to mucronulate (Fig. 7).Pringle 3525 (the type of E. digitata) possesses leaves that are ovate-lanceolate and acute (Fig. 10).Although the two collections are superficially rather different, and it is understandable why Watson named them as separate species, the numerous collections now available illustrate the thorough continuum in leaf shape and provide evidence that the two entities cannot be maintained as distinct.The leaves of this species are highly variable in shape (Fig. 7-10), and the types merely represent extremes in leaf morphology and are otherwise very similar.The only collection that I have seen with The plants in Morelos, Guerrero, and Oaxaca differ from plants throughout the northern range of this spe-ALISO cies in that they are densely glandular-pilose and consistently summer-fall annuals.The northern plants are nearly always perennial herbs or shrubs, and although frequently glandular-pilose, this is never to such an extent as in the southern plants.The significance of these differences is not yet apparent, and the plants from Morelos, Guerrero, and Oaxaca are here referred to Euphorbia subpeltata.
states that Sesse and another member of his expedition, Castillo, probably visited this area in August of 1792, and it is likely that the type was collected then.Modem collections of E. dioscoreoides ssp.dioscoreoides from the vicinity of Zitacuaro are known (e.g., Soto Nunez & Cortes A.2345 and Hinton et al. 13132).