444 FOSSIL PLANTS/Gymnosperms Characteristics Gymnosperms are a varied assemblage of plants whose principal botanical characteristic is reproduction by means of ovules, or seeds as they are called in their fertilized state This is an aspect of their biology that they share with the angiosperms One of the main differences between gymnosperms and angiosperms relates to the appendages associated with the ovules In the angiosperms, ovules are part of the flower One or more ovules is wrapped within a layer of tissue, the whole structure being called the carpel Flowers usually comprise carpels and additional organs such as pollen-bearing sacs, petals, and sepals Gymnosperms not have flowers The ovules are borne in cones or singly on more or less modified branches Furthermore, the ovules are not enclosed within a carpel, they are ‘naked’, hence the origin of the term gymnosperm (Greek gumnospermos: gumnos, naked ỵ sperma, seed) Modern gymnosperms are generally also woody shrubs and trees, but some have a vine-like growth form A more diverse array of growth forms, including groups with fern-like leaves, is known from the fossil record Ovules and seeds, wood, leaves, and pollen are frequently preserved as fossils, and together these provide our principal sources of evidence on the geological history of the group Classification Recent classifications of gymnosperms are based upon phylogenetic studies Because many aspects of gymnosperm phylogeny still remain unclear, modern classifications at the family or ordinal level must be regarded as very tentative There are problems in placing some extinct groups because they are still very poorly characterized In other instances, there are conflicts between the phylogenetic trees generated from comparative morphology and those produced from molecular systematic studies Despite these problems there is reason to be optimistic in the longer term The history of gymnosperm classification shows that better characterization of problematic fossils tends to lead to more stable ideas on relationships, and in the molecular world there is a seemingly inexhaustible supply of new data that should lead ultimately to a firm backbone phylogeny for the living groups Both approaches should converge on a mutually consistent and stable family tree One of the findings of recent phylogenetic studies is that some groups of gymnosperms are more closely related to angiosperms than they are to other gymnosperms (Table 1) In technical terms, gymnosperms are paraphyletic to angiosperms The precise nature Table Classification of seed plants (Spermatophyta) Spermatophyta ‘Calamopityales’ incertae sedisa, b Hyrdaspermalesa, b ‘Lyginopteridales’a, b Medullosalesa, b Euspermatales (taxa sedis mutabilis) Cycadales Callistophytalesa, b Coniferophytales Cordaitialesb Coniferales (Pinales) Glossopteridalesa, b Czekanowskialesb Ginkgoales ‘Peltaspermales’b ‘Corystospermales’ (‘Umkomastales’)b Caytonialesb Pentoxylalesb Bennettitalesb Gnetales Archaefructaceaeb Angiospermopsida ‘Gymnosperms’ ? Angiosperm Angiosperms Grouping is indicated through the use of indentation and the sequence a name occupies in the list Commonly applied names for individual groups are used without regard to Linnean rank order convention Questionably monophyletic groups are marked with quotes Sedis mutabilis indicates that groups which follow at the next level of the hierarchy are of uncertain relationship (i.e., form a polytomy or conflict strongly in different phylogenetic analyses) Incertae sedis indicates that the taxon bearing this epithet is of uncertain relationship to those at the same level Gymnosperms is a paraphyletic group It is equivalent to seed plants minus flowering plants (i.e., Spermatophyta minus Angiospermopsida) Modified after Crane and Kenrick (1997) Problems in cladistic classification: Higher level relationships in land plants Aliso 15: 87 104 a Groups typically included as Pteridosperms, or seed ferns, which are a heterogenous assemblage of gymnosperms with fern like foliage b Extinct group of this relationship, however, remains unclear, and this is one major source of conflict between family trees based on comparative morphology and those derived from molecular data Comparative morphology indicates that angiosperms are most closely related to living gymnosperms in the Gnetales and extinct gymnosperms in the Pentoxylales and the Bennettitales This has been dubbed the ‘anthophyte hypothesis’ (Figure 1) Molecular data tell a very different story They place Gnetales within Coniferales Angiosperms emerge as sister group to a monophyletic group comprising all living gymnosperms This has been dubbed the ‘gne-pine’ hypothesis (Figure 1) Neither hypothesis though is particularly strongly supported by the current data Furthermore, because molecular data cannot deal with the fossil groups, it is of course unclear where in molecular