C.—B

6

meadows. Happily, so far as foreign plants go, the forest is quite virgin, the elderberry, blackberry, and other dangerous intruders not having gained a footing in its recesses. 5. The Plant Formations. Plants are not scattered haphazard over a region, but occur in definite associations, which have been thus grouped in consequence of climate, the nature of the soil, the geological history of the land, the reaction of one plant upon another when brought into contact, and the presence of certain animals. Such groups of plants are called " plant formations," and these make up the vegetation of a region just as do the species the flora. To find out how such formations have originated, tracing their evolution, and studying the adaptations of their members to their surroundings, is one of the most important functions of botany, and has great bearings on agriculture, horticulture, and forestry. The first object of a botanical survey, then, is to classify the plant formations of a region. That having been done, details more or less complete can be furnished as to the adaptations of the species, and the causes which have led to the establishment of such formations as at present exist. Work of this kind is quite in its infancy in Australasia, and but a very few papers have as yet been published dealing in this manner with the New Zealand vegetation. Without going into the matter at any detail, it may easily be seen that the formations must be either primitive or modified, the modification having come about through the voluntary or involuntary action of man.* Actual primitive formations are more or less wanting in the temperate regions of the Old World, and it is the presence of such in New Zealand, and of so varied a kind, that makes this colony of extreme botanical interest to foreign scientists at the present time. Moreover, with the host of introduced plants, all kinds of stages and combinations exist in the evolution of modified formations. To examine and estimate these is one of the most urgent and yet difficult tasks for New Zealand botanists. On small islands, however, the problems are lesp complex than on a large continent, and so the study of the smaller New Zealand islands, such as the one under consideration, is of peculiar interest. The plant formations of Kapiti may be divided into —forest, coastal, shrubbery, meadow, and rock; and these again may be subdivided as need be. Since the forest is the formation par excellence upon which the welfare of the bird-life depends, it is treated at a greater length than are the other formations. A. The Forest. (a.) General Remarks. In an island so small and narrow as Kapiti, it might well be expected that the whole forest could be designated " coastal," but this is by no means the case. Coastal conditions, as I have attempted to show elsewhere,! rarely extend far from the shore, and they depend upon the presence of salt water, sea-spray, salt in the soil, special land-features such as dunce, and frequent winds from the sea. The forest of Kapiti, from its position, is exposed only in certain places to violent winds and usually to little sea-spray. It is therefore only in close proximity to the sea-shore that there is a coastal belt. Beyond this, right to the summit of the island, most of the forest species extend, but not in the same proportion. Also, there is distinctly a summit or an upper forest, as it may be called, which is in many respects different from that of the lower levels, and whose presence is due less to altitude than to a moister climate. Between the upper forest, the intermediate portion, and the coastal belt no hard-and-fast line, however, can be drawn. They merge gradually the one into the other. Thus the forest is here treated as one whole, and a certain amount of repetition avoided. Usually the Kapiti forest is very low, but it is a forest nevertheless, and not a scrub, if the former be defined as a collection of trees which have distinct trunks, even though these latter be short. Notwithstanding the above statement, in some places many of the trees become much more like shrubs, and a scrub-forest results. The forest is of the mixed type, and, notwithstanding what is said above, of a somewhat coastal character, owing to the presence of the kawakawa (Macropiper excelsum), the karaka (Gorynocarpus Iwvigata), and a species of daisy-tree (Olearia forsteri) in large quantities. But in many respects it differs materially from a typical New Zealand mixed forest, as is seen when treating of its physiognomy. (b.) Its Distribution. The distribution of the Kapiti forest is a striking example of the importance of wind as a factor in affecting the vegetation. As shown above, the northerly and southerly winds blow frequently with great violence, while the easterly wind is almost unknown. The consequence is that a dense green mantle of forest clothes all the eastern side of the island, where it has not been destroyed, from the sea to the margin almost of the high western cliffs, but on the north and south forest is wanting. As for this lack of forest in the south I can express no opinion, but in the extreme north it seems to me to have been always absent, or at best but a mere scrubby growth. Where on the eastern side meadow exists, simply means that the forest has been cleared, probably by burning, and that owing to the goats and sheep no new arborescent growth has been able to establish itself. To this latter statement is one probable exception, which is dealt with under the heading " Shrubbery," further on.

* Cockayne, L.: " A Short Account of the Plant Covering of Chatham Island " (Trans. N.Z. Inst., Vol. xxxi v, pp. 253 and 255; 1902). ■f See a paper entitled " Remarks on the Coastal Vegetation of the South Island of New Zealand," to appear in Trans. N.Z. Inst., Vol. xxxix, 1907.