First Devonian platyspermic seed and its implications in gymnosperm evolution

Abstract

IN the course of an investigation of fossil plants from the late Devonian rocks of Kiltorcan in southern Ireland obtained by re-excavation of the site earlier this year, we have discovered several seeds with features which contribute to the understanding of early gymnosperm evolution. The fossil flora of Kiltorcan has been known for more than a century1, and includes the well-known and widespread late Devonian genera Archaeopteris and Cyclostigma. In addition, Johnson2 briefly described and illustrated several oval bodies which he regarded as seeds, and to which he gave the name Spermolithus devonicus. His figures were at a small scale and rather uninformative, and most later workers discussing early evidence of seed plants have either been unaware of his claim, or make no reference to it. The only reference to Spermolithus known to us is Arnold's3 justifiable comment that “whether it is a seed or spore case is unknown to us”. In spite of the limitations of their state of preservation, we believe that the seeds we have found represent not only the earliest seed plants known from the British Isles, but also the earliest seeds with apparently platyspermic symmetry. In view of their closeness in age to the earliest generally accepted seed, Archaeosperma, of radiospermic character, we believe that this favours a more or less synchronous origin of radiospermic and platyspermic gymnosperms. This suggests that these two seed lineages have been separate since their inception, rather than having had a single evolutionary origin followed by rapid divergence within the Carboniferous.