Organography and vascular anatomy of the flower and certain aspects of embryology are reported for eight members of the family Diapensiaceae: Diapensia lapponica, D. lapponica subsp. obovata, Galax aphylla, Pyxidanthera barbulata, P. brevifolia, Schizocodon soldanelloides, Shortia galacifolia, and S. uniflora. 2. The flowers are actinomorphic, essentially hypogynous, and pentamerous except for the tricarpellate pistil. There are five fertile stamens opposite the sepals, and, in all but Diapensia and Pyxidanthera, five staminodes opposite the petals. The stamens and, to a varying extent, the staminodes are epipetalous except in Galax in which the petals are almost completely free and the stamens and staminodes monadelphous. Differences occur in length, orientation and number of pollen sacs, plane of dehiscence, and occurrence of appendages. Placentation is basically axile but may approach parietal in the top of the ovary. Ovules are few to many. 3. The sepals receive three traces in most species, but in Shortia galacifolia the number may be as high as seven in the lower and outermost of the helically arranged sepals, or there may be a single trace in an occasional sepal or in all sepals of an occasional flower of Pyxidanthera (particularly P. barbulata). Occasionally the lateral traces may be independent or diverge from the median trace, but usually the laterals of adjacent sepals arise from a single gap and diverge in the outward course of the common bundle. The single petal traces may have a common origin with the joint lateral sepal strands or be independent close above them. Each stamen receives a single trace, which in some species diverges from the median sepal strand and in others arises independently above the closure of the median sepal gap. This strand terminates in the connective without any bend. The vascular supply to the carpels is variable. The three dorsal traces vary in origin and in whether or not they branch; they traverse the style. There may be one to several septal bundles. There are three ventral bundles in the planes of the locules or of the septa except in Shortia where there are six. Bundles in the planes of the septa split and supply half-placentae of adjacent carpels. 4. The stamens are characterized by having anthers with a well-developed endothecium, a glandular binucleate tapetum, single pollen grains that are probably two-celled at maturity, and lengthwise dehiscence-either longitudinal, oblique, or transverse depending on the orientation of the pollen sacs. 5. In the frequently anatropous ovule (from hemitropous to campylotropous depending on species or position on the placenta) a single archesporial cell (in Diapensia often two or more) gives rise to the megaspore mother cell. Meiosis results in a linear tetrad (T-tetrads occasional in Diapensia) of which the chalazal spore normally functions. Frequently in Shortia galacifolia and occasionally in Diapensia a spore other than the chalazal one is functional. Three mitotic divisions result in the usual eight-nucleate gametophyte. Development is thus of the Polygonum type. The antipodals multiply in both species of Shortia, and the polars fuse in all species. The single-layered nucellus disintegrates during development so that the gametophyte lies in contact with the inner layer of the single integument. No endothelium is differentiated. 6. Development of the endosperm, which precedes that of the embryo, is cellular. The first division is usually transverse, but may be longitudinal in Shortia galacifolia. The second division is normally longitudinal in the micropylar cell and, in Diapensia and Galax, transverse in the chalazal cell. In Shortia, however, it is normally longitudinal in the chalazal cell also. No haustoria, micropylar or chalazal, are formed, but the antipodals persist through the early stages of seed development. The zygote elongates only slightly before undergoing a transverse division. 7. The floral morphological and embryological features of the Diapensiaceae are compared with those of the Ericales in general. The number of differences, particularly in embryology, suggests that the Diapensiaceae should be excluded from the Ericales.
No takes yet. Share an insight, caveat, or question.
Barbara F. Palser (1963) studied this question.
Synapse has enriched 5 related papers. Consider them for comparative context: