We review the diversity and development of archegonia, the female reproductive organs of land-plant gametophytes. The archegonium is a uniquely land-plant structure, and studies of its evolution benefit from use of a comparative approach in a phylogenetic context. Archegonia of most land plants share a common developmental motif, here termed a T-shaped pattern. A primary axial cell produces a primary cover cell and a central cell by horizontal division. The upper cell usually divides vertically and the lower one horizontally. In mosses such as Atrichum, the T-shaped stage is shifted towards the end of archegonium development, whereas in vascular plants it appears at the beginning of development, but these stages are still probably homologous. The fully exposed archegonia are traditionally viewed as an ancestral (plesiomorphic) condition in land plants, but there is no direct support for this view. We speculate that the fully exposed condition is derived and synapomorphic for setaphytes (mosses and liverworts). The fully sunken hornwort archegonia may be similar to the ancestral type of land-plant archegonia. Developmental evidence suggests that archegonium necks of setaphytes and tracheophytes are not homologous to each other. The neck wall of pteridophytes is composed of four-celled tiers, and one such tier is present in gymnosperms with motile male gametes. Neck-cell arrangement is much more plastic in archegonia of gymnosperms with sperm cell delivery by pollen tube (siphonogamy), in which the neck plays a role similar to pollen-tube transmitting tissue of angiosperms. Angiosperm synergids are probably homologues of gymnosperm neck cells, and the angiosperm egg cell is probably homologous to the ventral canal cell of gymnosperms. Developmental genetic bases of archegonium diversity in land plants remain to be understood. Even descriptive developmental data are currently missing or controversial for some key lineages of land plants.
Thismia is characterized by an exceptionally complicated floral morphology that is currently not understood properly. In the taxonomic literature, descriptive rather than morphological terms are often applied to parts of the flower in Thismia, relating to the general appearance of the floral organs instead of their precise homologies. Precise understanding of the floral structure is complicated by the rarity of Thismia spp. and the paucity of appropriate material. Here we provide a comprehensive study of reproductive organs of three Thismia spp. (T. annamensis, T. javanica and T. mucronata) including the first investigation of inflorescence architecture and early floral development in Thismiaceae. We found a hitherto unknown diversity of the reproductive shoots in the genus, manifested in the number of floral prophylls (two or three, in contrast to a single prophyll in the vast majority of monocots) and in the branching plane resulting in two distinct inflorescence types, a drepanium and a bostryx. We report the non-acropetal sequence of initiation of floral whorls (with stamens being the last elements to initiate), never previously described in monocots, and the gynoecium composed of completely plicate carpels, also a rare feature for monocots. Floral vasculature is relatively uniform in Thismia, but significant interspecific differences are found in tepal innervation, including the number of tepal traces; some of these differences are not immediately related to the external tepal morphology. We argue that the annulus, which acts as a roof of the hypanthium, possesses an androecium nature and represents congenitally fused bases of stamen filaments. We describe the stamens as laminar structures, which are also shortly tubular in the distal part of the supraconnective with the adaxial tubular side forming a skirt-like appendage. Finally, the placentas, which are column-like when mature, are initially parietal, becoming secondarily similar to freecentral placentas through schizogenous separation from the ovary wall.
Finding morphological differences between cytotypes that are stable throughout their geographical range is important for understanding evolution of polyploid complexes. The ancient monocot lineage Acorus includes two groups, of which A. calamus s.l., an important medicinal plant, is a polyploid complex with a centre of diversity in Asia. European plants are sterile triploids introduced by humans. An early study suggested that plants from temperate Asia are tetraploids, but subsequent work revealed diploids and triploids rather than tetraploids in Asiatic Russia; however, cytotype diversity in Western Siberia is insufficiently known. We document the occurrence of diploids and triploids in Western Siberia. Triploids that do not differ in genome size from European Acorus are abundant in the valley of the river Ob where the ability for extensive vegetative propagation provides ecological advantages. An isolated population of aneuploid triploids with 33 chromosomes is found outside the Ob valley. Flow cytometry provides an efficient tool for identification of aneuploid plants in Acorus. All triploids are sterile, but their flowers develop uniform parthenocarpic fruits. Fruits of diploids usually vary in size within a spadix depending on the number of developing seeds. In contrast to North America, where the native diploid plants differ from the introduced triploids by the absence of a secondary midrib of the ensiform leaf blade, Siberian diploids are similar to triploids in possessing a secondary midrib. We confirm that diploids differ from triploids in the size of air lacunae in leaves, which is determined by cell number rather than cell size in septa of aerenchyma. A combination of spathe width and spadix length measured after the male stage of anthesis shows different (slightly overlapping) patterns of variation between diploids and triploids in our material.