Early land plants reproduced in the fashion of ferns: spores germinated into small gametophytes, which produced sperm. These would swim across moist soils to find the female organs (archegonia) on the same or another gametophyte, where they would fuse with an ovule to produce an embryo, which would germinate into a sporophyte.[75] Heterosporic organisms, as their name suggests, bear spores of two sizes microspores and megaspores. These would germinate to form microgametophytes and megagametophytes, respectively. This system paved the way for seeds: taken to the extreme, the megasporangia could bear only a single megaspore tetrad, and to complete the transition to true seeds, three of the megaspores in the original tetrad could be aborted, leaving one megaspore per megasporangium. The transition to seeds continued with this megaspore being "boxed in" to its sporangium while it germinates. Then, the megagametophyte is contained within a waterproof integument, which forms the bulk of the seed. The microgametophyte a pollen grain which has germinated from a microspore is employed for dispersal, only releasing its desiccation-prone sperm when it reaches a receptive megagametophyte.[32] Lycopods go a fair way down the path to seeds without ever crossing the threshold. Fossil lycopod megaspores reaching 1 cm in diameter, and surrounded by vegetative tissue, are known these even germinate into a megagametophyte in situ. However, they fall short of being seeds, since the nucellus, an inner spore-covering layer, does not completely enclose the spore. A very small slit remains, meaning that the seed i

still exposed to the atmosphere. This has two consequences firstly, it means it is not fully resistant to desiccation, and secondly, sperm do not have to "burrow" to access the archegonia of the megaspore.[32] A middle Devonian precursor to seed plants from Belgium has been identified predating the earliest seed plants by about 20 million years. Runcaria, small and radially symmetrical, is an integumented megasporangium surrounded by a cupule. The megasporangium bears an unopened distal extension protruding above the mutlilobed integument. It is suspected that the extension was involved in anemophilous pollination. Runcaria sheds new light on the sequence of character acquisition leading to the seed. Runcaria has all of the qualities of seed plants except for a solid seed coat and a system to guide the pollen to the seed.[83] The first "spermatophytes" (literally:seed plants) that is, the first plants to bear true seeds are called pteridosperms: literally, "seed ferns", so called because their foliage consisted of fern-like fronds, although they were not closely related to ferns. The oldest fossil evidence of seed plants is of Late Devonian age and they appear to have evolved out of an earlier group known as the progymnosperms. These early seed plants ranged from trees to small, rambling shrubs; like most early progymnosperms, they were woody plants with fern-like foliage. They all bore ovules, but no cones, fruit or similar. While it is difficult to track the early evolution of seeds, the lineage of the seed ferns may be traced from the simple trimerophytes through homosporous Aneurophytes.