The Importance of Fossils in Elucidating Seed Plant Phylogeny and Macroevolution

Doyle, J.A. and Donoghue, M.J., 1987. The importance of fossils in elucidating seed plant phylogeny and macroevolution. Rev. Palaeobot. Palynol., 50: 63--95. In order to gain insights on the controversial question of the value of fossils in understanding phylogeny and macroevolution, we used numerical parsimony methods to analyze a data set amassed for a cladistic study of living and fossil seed plants, variously modified by subtraction and re-addition of fossil groups. Some cladograms based on extant groups alone are consistent with relationships derived from the whole data set, but the direction of leaf and sporophyll evolution in seed plants and floral evolution in angiosperms and Gnetales is equivocal. However, use of Carboniferous seed ferns as outgroups favors the concept that all seed plants were derived from ancestors with fernlike leaves, that there was a double trend to linear-dichotomous leaves in coniferopsids and Gnetales, and that gnetalian flowers are reduced relative to those of angiosperms and Bennettitales, as inferred from the whole data set. With the whole data set, it is slightly more parsimonious to assume that coniferopsids were derived from Callistophyton-like platyspermic seed ferns than from Archaeopteris-like progymnosperms, but if Archaeopteris were unknown the seed fern hypothesis would be strongly favored, while if Callistophyton were unknown both hypotheses would be equally parsimonious. Bennettitales strengthen relationships between Gnetales and angiosperms, which are only slightly stronger than links between Gnetales and coniferopsids when extant data alone are considered, while Caytonia clarifies reproductive homologies of angiosperms, Bennettitales, and Gnetales and their origin from platyspermic ancestors. Although fossil information does not radically alter inferred relationships among extant taxa in seed plants, it may in other groups, and it provides unique evidence on the sequence of events and possible adaptive factors involved in the origin of groups. Both contributions are especially important when there has been extensive homoptasy and/or when living groups are isolated from each other by large gaps, as is often true of higher taxa.