Pseudopanax crassifolium, Suttonia australis, Myrtus pedunculata and Wintera colorata. Also present in most of the communities are Myrtus obcordata, Nothopanax Edgerleyi and N. simplex. The shrublayer is composed of young plants of the foregoing species together with Coprosma foetidissima, C. propinqua, C. rhamnoides and C. rotundifolia. Seedling or juvenile plants of Nothofagus are extremely scarce and tree-ferns are generally absent or few in number, though the prostrate Alsophila Colensoi is sometimes present. Ferns that are common as epiphytes are Asplenium flaccidum, Polypodium diversifolium and Cyclophorus serpens, this last ascending the trees to the topmost branches. The principal ground-species are Blechnum discolor (often dominant), B. procerum, B. fluviatile, B. lanceolatum, Polystichum vestitum, and Astelia nervossa. Rubus australis, Parsonsia heterophylla and Clematis indivisa occur occasionally, and in most of the localities lichens, mosses and liverworts are much in evidence. The conditions met with on the forest-floor in all the above associations are similar, a thick undecomposed layer of fallen leaves, a mass of interlaced rootlets very near the surface, and regeneration of Nothofagus practically non-existent under the canopy of the associations. The formation of the leaf-litter is probably due to the dense shade depressing the temperature of the forest-floor, thus slowing down the biological and chemical processes in the soil and eventually leading to humus accumulation. It must be remembered in this connection that leaves of different species vary greatly in their power to resist decomposition. What effect the high soil acidity in itself has on the regeneration of seedling beech is difficult to estimate without direct experiment. Other factors, such as lack of light, leaf-litter and humus-accumulation, certainly play a part, and it is interesting to note that investigations by Hesselman on the conifers in North Sweden show that seedlings appeared and became established there only where the soils were nitrifying. It is probable that the nature of the original rock from which the soil has been derived has some effect on its hydrogen-ion concentration; but as far as our observations go, once forest or other plant-formations have been in existence on any area for long periods of time, the nature of the underlying rock generally does not seem to have any great effect on the surface soil-reaction, old established forest, for instance, making in the main its own soil conditions. It is noteworthy that lack of regeneration in N. Menziesii forests is associated with poor light conditions, accumulation of leaf-litter and marked acidity of the soil, while regeneration takes place where there is an abundance of light, absence of leaf-litter and where the soil is, in the majority of cases, much nearer neutrality. Rain-forest Communities. The next class examined were soils of the various rain-forest associations present in the botanical area. These, unlike the Nothofagus communities, showed no uniformity, but were characterized by values that fluctuated in a remarkable manner, wide differences in hydrogen-ion concentration being observed in quite small areas;