Ovules and seeds of Barbeya with additional arguments for an urticalean affinity of the Barbeyaceae

Abstract

SUMMARY Barbeya oleoides has a bitegmic, crassinucellate and anatropous ovule. The inner and outer integuments are about five layers thick and not vascularized. The endosperm is initially nuclear, to become cellular later. The mature seed coat is unspecialized, remains parenchymatic and is locally compressed, except for the exotesta and the tanniniferous endotegmen. The exotesta is perforated by distinct crateriform holes. The embryological and anatomical seed coat characters support an urticalean affinity of the Barbeyaceae.

Urticalean rosids: circumscription, rosid ancestry, and phylogenetics based on rbcL, trnL‐F, and ndhF sequences

“…Likewise, this placement is consistent with Berg's (1989) conclusion that Barbeyaceae are not closely related to any of the urticalean families and that their placement with the urticalean rosids has persisted not because of shared characters, but rather due to the lack of any clear alternative placement. However, Bouman and Boesewinkel (1997) argued that seed characteristics of Barbeya indicate an undisputed urticalean affinity. Of the other members of the sister group to urticalean rosids, Elaeagnaceae and Rhamnaceae have been suggested previously as putative relatives (Hallier, 1905;Thorne, 1968).…”

“…A complex of Rhamnaceae, Elaeagnaceae, the monotypic Barbeyaceae (Arabia and northeastern Africa; see Bouman and Boesewinkel, 1997;Thulin et al, 1998), and the enigmatic, monotypic Dirachma (Socotra, south of Yemen; previously placed in Geraniaceae but more recently in its own family; see Boesewinkel and Bouman, 1997) comprises, at least in part, the sister group to the urticalean clade. The placement of Barbeya outside urticalean rosids and in a complex of taxa including Dirachma, Elaeagnaceae, and Rhamnaceae is consistent with recent molecular analyses (Thulin et al, 1998;Richardson et al, 2000;Savolainen et al, 2000a, b;Soltis et al, 2000) and with the suggestion of Tobe and Takahashi (1990), based on trichome and pollen morphology, that Barbeyaceae are sufficiently different to be removed from the urticalean clade.…”