The Tree-Ferns of the genus Cyathea in Australasia and the Pacific

The present account of this genus follows the lines of that adopted for Flora Malesiana (ser. 11, 1, part 2, 1963). While studying Malesian species, I examined the types of those in neighbouring regions, to discover to what extent Malesian species were distributed further to the East and South-east. I found very few species with such extended distribution, but it was evident that the species of Australasia and the Pacific are closely related to those of Malesia. There has been no attempt at a full comparative survey of all the species of Cyathea over this vast area since Hooker and Baker’s Synopsis Filicum (2nd edition, 1874). Later accounts have been very summary, or confined to limited parts of the region, and there has been inconsistency as between dilferent accounts in the interpretation of some specific names, especially those originating with Forster. I have now examined type material of almost all species, and hope that I have resolved most of the discrepancies of interpretation. A few new species are also here described. Though the scales of the stipe provide characters by which any species may be placed in its subgenus, other characters are usually necessary for distinguishing individual species. Working from herbarium specimens, I find that the only way to distinguish clearly between species is to examine both indusia (if any) and scales on the lower surface of leaflets with a binocular dissecting microscope at a magnification of 25, and in the descriptions I have attempted to state concisely the distinctive characters of scales and indusia thus seen. The only previous authors who described scales at all carefully were Mettenius and Christensen. The indusia need even more careful examination than the scales. Some which have been described as cup-shaped are in fact hood-shaped, being open on the side towards the margin of the leaflet (e.g. C. cunninghamii); at the other extreme are indusia so small that they are covered by the ripe sporangia and so have sometimes been overlooked (e.g. in C. decurrens).