WMTTEN ET AL. Relationships of Zygopetalinae and not slit-like. Dressier (1993), who included Vargasiella in Zygopetalinae, mentioned possible placement in its own subtribe; Romero and Camevali (1993) validated the subtribe Vargasiellinae, which was also recognized by Szlachetko (1995). We were unable to obtain extractable material of this genus for inclusion in this study and its placement remains uncertain. Our sampling within the Huntleya grade is more complete, especially for Central American taxa, and some conclusions and taxonomic transfers are justified by the analyses. The discussion is arranged by the genera recognized in Figs. 4 & 5, although Hoehneella is not in the figure (see discussion below). Cryptarrhena This genus (two species) is morphologically anomalous within the subtribe and is isolated on a very long branch. Its placement within the subtribe in the combined cladogram is unresolved, but it is sister to the Huntleya clade in many of the shortest trees. The spicate, pendent inflorescences have numerous, small flowers, whereas most of the Huntleya clade have single-flowered inflorescences. Cryptarrhena lunata has fleshy, strongly keeled leaves and lacks pseudobulbs, whereas C. guatemalensis has thinner leaves and small pseudobulbs. Nevertheless, several characters link Cryptarrhena to other genera of Zygopetalinae. The anchor-shaped lip is similar to that of Dichaea, and the column bears a conspicuous clinandrium (hood) similar to those of Huntleya and Chaubardia, and a distinct basal tooth. Within the subtribe, many-flowered inflorescences are also present in Galeottia, Warrea, Warreopsis, and Zygopetalum, among others. Chaubardia Florally, Chaubardia (three species) is very similar to Huntleya; both possess flat, open flowers with rhomboid lips bearing a conspicuously toothed callus. The columns of both genera possess lateral wings and often a hooded clinandrium, but the sepals and petals are narrower than those of Huntleya. Chaubardia is characterized by small, inconspicuous pseudobulbs at the base of fan-shaped growths, whereas Huntleya species lack pseudobulbs. The molecular data strongly support monophyly of Chaubardia and its separation from Huntleya. Hoehneella We were unable to obtain extractable material of this small genus of 1 or 2 species. Morphologically, it is similar to Huntleya and Chaubardia. According to Senghas and Gerlach (1992-1993), the plants possess small pseudobulbs similar to those of Chaubardia and its viscidium is transversely elliptic and lacks a stipe. [Type: H. gehrtiana (Hoehne) Ruschi] Huntleya This is a distinctive and easily recognized genuswithabout 13 species. Theplantslackpseudobulbs, and some species possess elongate rhizomes separating the fan-shaped growths. The flowers are large, star- shaped, and flat with relatively broad sepals and petals and are probably fragrance-reward flowers pollinated by male euglossine bees. Dichaea Dichaea is the largest and most distinctive genus in the subtribe (about 111 species) and is widely distributed from Mexico to Brazil. It is characterized by long, many-leaved, pseudomonopodial stems that are pendent in many species. Solitary flowers bearing an anchor-shaped lip are produced successively along the leafy stems, and all species are probably pollinated by fragrance-collecting male euglossine bees (although autogamous forms occur). The genus is monophyletic and the representatives of Dichaea are on a relatively long branch and are remarkable for the high levels of sequence divergence among the species; the branch lengths within Dichaea are greater than the lengths among most genera within the subtribe. These data indicate that sequencing of ITS and plastid regions has great potential for resolving species relationships within this moderately large genus. Conversely, the low levels of sequence divergence within many genera (e.g., Kefersteinia, Bollea, Pescatorea) indicate that sequencing these regions for additional taxa will not help to clarify relationships within these genera. The remaining taxa within the Huntleya clade (Fig. 4) include many species that have been included in Chondrorhyncha. Generic delimitation within this clade has been difficult and many species have complex nomenclatural histories as orchidologists have shifted them from one genus to another. This taxonomic confusion probably reflects repeated evolutionary changes in pollination mechanisms that produced homoplasious floral morphologies. Many species in this clade produce gullet flowers that appear to be nectar deceit flowers for long-tongued visitors, probably nectar-foraging euglossine bees (Ackerman 1983). The funnel-shaped lips do not produce a true spur, but do have a notch on either side of the base of the lip that permits passage of a bee's tongue. The lateral sepals are swept back and revolute, forming a tubular false spur LANKESTERIANA5(2):