The ultimate basis of relationship among plants is their origin, and the known facts of history should take precedence over superficial resemblances in life cycles that are often the results of parallel development. The larger categories such as divisions and classes should be founded upon groups of characters expressed by vegetative as well as reproductive structures. The fossil record indicates that the 2 main gymnosperm lines, the Cycadophyta and the Coniferophyta, are distinct as far into the past as they can be traced, and are only very remotely, if at all, related. The naked seed is apparently an independent development within these 2 groups and is not an indication of affinity. The term "Gymno-spermae" is therefore inappropriate as a categorical designation and its use should be discontinued. A modified classification of the Pteropsida is proposed in which the division consists of (1) the Pteridophyta, restricted to the true ferns; (2) the Cycadophyta, embracing the pteridosperms, cycads, and cycadeoids; (3) the Coniferophyta, which includes the Cordaitales, Ginkgoales, Taxales, and the Coniferales; (4) the Chlamydosperm-ophyta, consisting of the Ephredales and the Gnetales; and (5) the Angiospermophyta.