Along with the origins of vascular plants and seed plants, the origin of flowering plants (angiosperms) represents one of the three most significant evolutionary radiations of land plants during the last 450 million years. With over 250,000 extant species, angiosperms are the largest and most diverse group of plants to have evolved. Paradoxically, we know less about the origin and early evolutionary history of flowering plants than we do about many considerably older groups of land plants. Indeed Darwin's "abominable mystery" of the origin and early evolution of flowering plants, continues to challenge evolutionary biologists. For 100 years, a "double fertilization" process that yields a diploid embryo and a triploid embryo-nourishing tissue called endosperm has been considered a defining feature of angiosperms. Although endosperm is known to be a ubiquitous feature of angiosperms, the participation of a sperm in a second act of fertilization to produce endosperm has been carefully documented in only a relatively small number of derived monocot and eudicot taxa. It is clear from recent phylogenetic analyses of basal angiosperms (and a survey of the plant fertilization biology literature) that a second fertilization event, long assumed to be an important and defining feature of flowering plants, has yet to be conclusively documented and characterized in any putatively basal angiosperm. This presents an enormous gap in our knowledge of the most basic aspects of fertilization biology that characterize "primitive" angiosperms. The proposed research seeks to determine whether a second fertilization event is associated with the initiation of endosperm tissue in basal angiosperms; or whether endosperm tissue in some basal angiosperms is strictly maternal in origin. Modern cell biological approaches will be used to study and document the presence or absence of double fertilization in a broad cross section of basal angiosperm lineages. If it is discovered that double fertilization is not a synapomorphy of angiosperms (and is instead a synapomorphy of a subset of angiosperms), this will have a significant impact on hypotheses of endosperm evolution, and ideas concerning the possible adaptive value (key innovation status) of this unique embryo-nourishing tissue in flowering plants. A second component of the proposed research seeks to determine the developmental basis for the diversification of endosperm developmental patterns during the early evolutionary radiation of flowering plants. Three typological categories have been used to circumscribe endosperm structure (e.g. free nuclear, cellular, helobial) for the past century. However, there has never been an analysis of the developmental basis for transformations of one pattern into another. Thus, little is known about key character transitions and the associated modifications of basic aspects of development (i.e. relationship of mitosis and cytokinesis, polarity, symmetry, and cell autonomous fates). Study of endosperm development across the phylogenetic spectrum of basal angiosperms will lead to new paradigms for the interpretation of the diversification and evolution of endosperm. A concerted effort to understand comparative and evolutionary aspects of fertilization biology and endosperm development in "primitive" flowering plants has never been before been attempted. As a consequence, virtually nothing is known of the early evolution and diversification of patterns of reproduction among angiosperms. This comparative analysis of reproductive features from key basal angiosperm lineages will address important issues of character distribution and evolution associated with the remarkable evolutionary radiation of flowering plants; and will mark the first explicit attempt to circumscribe the most basic aspects of fertilization biology and endosperm development in the earliest angiosperms.