Ancestral traits and specializations in the flowers of the basal grade of living angiosperms

Abstract

New morphological and phylogenetic data prompt us to present an updated review of floral morphology and its evolution in the basal ANITA grade of living angiosperms, Chloranthaceae, and Ceratophyllum. Floral phyllotaxis is complex whorled in Nymphaeales and spiral in Amborella and Austrobaileyales. It is unresolved whether phyllotaxis was ancestrally whorled or spiral, but if it was whorled, the whorls were trimerous. The flowers are probably ancestrally bisexual because in most families with unisexual flowers these flowers exhibit rudiments of the opposite sex. Carpels are largely ascidiate and the closure line is short, either transverse or longitudinal. A style is usually absent or, if present, generally short and not plicate. Angiospermy (carpel sealing) is by secretion, rather than postgenital fusion, except in large‐flowered Nymphaeales and in Illicium, correlated with unusual fruits. Carpels with a single, median, pendent ovule are probably plesiomorphic. Chloranth‐ aceae and Ceratophyllaceae have an unsettled phylogenetic position, but in some phylogenetic analyses they form a clade, which may be sister to the remaining mesangiosperms (Magnoliidae, monocots, eudicots). This position is supported by their carpel characters, which are similar to those of the ANITA grade and different from those of most other mesangiosperms.

Pseudoasterophyllites cretaceus from the Cenomanian (Cretaceous) of the Czech Republic: A possible link between Chloranthaceae and Ceratophyllum

“…Pseudoasterophyllites still had bracts subtending the stamens, but these were lost in the common ancestor of the Pennipollis plant and Ceratophyllum (and independently in the Hedyosmum line), and anther dehiscence was modified from latrorse to introrse or extrorse. Further specializations in Ceratophyllum were loss of the pollen aperture and extreme reduction of the exine (Takahashi, 1995); origin of its peculiar style, which is larger on the presumed ventral side of the carpel (Endress, 1994;Iwamoto & al., 2003;Endress & Doyle, 2015); and dry fruit wall (fleshy in the Pennipollis plant, unknown in Pseudoasterophyllites). Because vegetative morphology of the Pennipollis plant is unknown, it is equivocal whether the linear leaves of Pseudoasterophyllites and the dissected leaves of Ceratophyllum are autapomorphies of these genera or originated earlier.…”

“…4B, 5C, F). We interpret this ring as the chalazal attachment scar (hilum); its asymmetry is typical of orthotropous ovules that are attached near the apex of a uniovulate carpel locule, as in Amborella, Chloranthaceae, and Ceratophyllum (Endress, 2011;Endress & Doyle, 2015). At the opposite end, the seed is often broken (Figs.…”

Was the ancestral angiosperm flower whorled throughout?

“…* Based on perianth in Hedyosmum. ** Following interpretation of flowers as naked unicarpellate or unistaminate (Endress and Doyle, 2015).…”

Cecilanthus polymerus, a novel multiparted flower from the mid-Cretaceous Rocky Point locality, Maryland

“…Because the number of tepals, stamens, and carpels per whorl is high (each organ is many times narrower than the circumference of the floral base at the level of attachment of the organs), it is likely that the whorls are in fact part of a complex-whorled system, in which the number of parts per whorl increases in successive whorls by production of parts in double positions (Staedler and Endress 2009;Endress and Doyle 2015). However, because the basal part of the flower with the attachment zone of the outermost whorls is not preserved, the onset of such whorl complexity cannot be reconstructed.…”