Art. XXXVIII.—Notes on the Vegetation of the Open Bay Islands.

Transactions and Proceedings of the Royal Society of New Zealand, Volume 37, 1904, Page 368

Art. XXXVIII.—Notes on the Vegetation of the Open Bay Islands. By L. Cockayne, Ph.D. Communicated by A. H. Cockayne. [Read before the Wellington Philosophical Society, 2nd November, 1904.] Plate XXIII. 1. General Remarks. Lying in the Tasman Sea, some three nautical miles from the shore of south Westland, almost opposite the mouth of the Okuru River, are the two small islets known as the Open Bay

Islands. From their structure, and from what is known regarding the geological history of New Zealand, they evidently at no very distant date formed a portion of the mainland. It seems, then, instructive from a phytogeographical point of view to give a brief sketch of their vegetation, at the same time comparing it with that of the adjacent coast, so that the effect of isolating such coastal vegetation upon a gradually diminishing land-area and exposing it in consequence to slowly increasing unfavourable conditions may be estimated. The importance of this study is also very much enhanced from the fact that the vegetation of the islands is quite in its primeval condition, there having been no settlement of any kind; and, so far as I could observe, with the exception of a few plants of Poa annua, introduced plants are absent. Now, the coast of Westland, thanks to the excessive rainfall of more than 254 cm. yearly, is covered almost to high-water mark with a dense forest, made up of numerous species of plants, but which differ in numbers and grouping according to the drainage conditions of the ground. Such a forest is, however, a typical evergreen rain-forest, and contains many epiphytes, lianes, tree-ferns, mosses, and liverworts. As for the lianes, the tree-trunks are frequently so thickly covered with climbing species of Metrosideros or Freycinetia banksii that their bark is quite hidden. With forest similar to the above would the Open Bay Islands be covered when forming part of the mainland. During a trip in the Government steamer “Hinemoa” I had the rare opportunity of visiting these islands, in February, 1903, and it is from notes hastily taken during that visit that this paper is compiled. The islands are two in number, divided from each other by a narrow strait. The larger and more northerly of the two is perhaps at most half a mile in length. On the east its surface terminates abruptly in precipitous cliffs, ranging in height from 15 m. to 30 m., but towards the west it gradually slopes to the sea. To the north and west are many flat rocks, which are left quite bare at low water. At the southern extremity this island is cut into two unequal parts by a deep chasm, through which the sea dashes at high water. The smaller island consists of a low conical hill 45 m. or thereabouts in height. Both islands are formed of limestone, which is overlaid by a rather deep covering of soil, consisting of coarse, dark-coloured peat, containing a considerable percentage of imperfectly decayed vegetable matter. This soil is extremely loose, both from its texture and from being honeycombed with the holes of mutton-birds. It is also saturated with water, so that, when

dug into, moisture from all sides flows into the newly-made cavity. At the bottoms of the mutton-birds' holes, too, water must sometimes lie.* This is the station where the leech recently described by Dr. W. B. Benham under the name Hirudo antipodum was discovered by Mr. W. Dunlop and myself. (See Trans. N.Z. Inst, vol. xxxvi., p 185, 1904.) At the southern extremity of the northern island is the remains of an old beach, or river-terrace, built up of rounded stones firmly cemented together, none of which contain lime. This formation Mr. G. J. Roberts, Chief Surveyor for Westland, informs me is of glacial origin, and corresponds to similar deposits on the mainland. Both islands are densely covered with vegetation, and the flat rocks mentioned above contain many deep pools in which seaweeds are abundant, while the rocks themselves are thickly covered with Hormosira banksii. 2. Vegetation of the Larger Island. (a.) Plants of Rocks. The richness of plant life on the rocks depends chiefly on their position with regard to sun and wind, and upon the steepness of the rock-face. Where the cliffs are in the shade, and consequently always more or less wet, they are densely clothed with vegetation; but in other places the plant-formation is an open one, the plants occurring in larger or smaller patches, while in others again the rocks are bare, except for lichens. Over the top of the precipices† Here is the breeding-ground of the spotted shag (Phalacrocorax punctatus), large numbers of young being on the cliffs and adjacent rocks at the time of our visit. facing the mainland hangs Veronica elliptica. Below this belt is frequently a zone of Asplenium obtusatum, beneath which are long vertical breadths of Tillæa moschata and Lepidium oleraceum mixed together. Samolus repens occurs on the drier portions of the rocks, occupying crevices, and thus sometimes making long lines more than 2 m. in length and some 17 cm. in breadth. Finally, in some places the thick-leaved fern Lomaria dura grows more or less luxuriantly. On the flat rocks at the northern end of the island are Samolus repens, Apium prostratum, Scirpus nodosus, young plants of Veronica elliptica, Phormium tenax at the margin of the rocks coming almost to the water's edge, and in some places are masses of Rumex flexuosus with shoots attaining a length of ± 1 m. (b.) Zonal Distribution of Plants on Surface of Island. As seen from the above description of rock-vegetation, the distribution is in part zonal, but the surface of the island, especially

that of the smaller southern portion, shows such distribution in a more striking manner. Commencing near the edge of the cliffs, or in some places extending on to the cliffs themselves, is a zone of Tillæa moschata; sometimes quite pure, or in other places intermingled with Samolus repens and Apium prostratum. The Tillæa is in certain situations, especially where the ground is particularly wet, of great luxuriance—the stems, e.g., about 32 cm. in length, with the lower half prostrate and bare, but conspicuous through its red colour, while the upper half is erect and covered thickly with short succulent green leaves. In some places Tillæa may be wanting, in which case the continuity of the zone is preserved by breadths of Lepidium oleraceum, with spreading shoots ± 72 cm. in length. The “Tillæa zone” is succeeded by a second one consisting of Carex comans, 4 m. or more in width, the individual plants forming pale-coloured tussocks 17 cm. tall, and with a spread of 32 cm. The third zone, about 9 m. in breadth, consists of the very common maritime fern Asplenium obtusatum, growing densely and with the thick coriaceous fronds 44 cm. tall. These arise from stout creeping rhizomes, ± 30 cm. in length by ± 22 cm. in circumference—i.e., if the bases of the fallen fronds be included—the actual rhizome measuring only ± 3.5 cm. × ± 3 cm. Such large breadths of dense fern, with the fronds of a uniform height, and all touching, make a remarkable sight (see Plate XXIII.). As showing the looseness of the soil in which this fern grows, pieces of rhizome were pulled up with peat attached measuring 73 cm. × 50 cm. Succeeding the zone of pure Asplenium obtusatum is in some places a fourth zone, of which a well-marked variety of Veronica ellliptica is occasionally the principal plant, or this may be mixed with or replaced by Muehlenbeckia adpressa, combined with certain other plants. The Veronica elliptica is 4 m. or more in height, the main trunk is prostrate at the base, then ascending and spreading with straggling branches. This variety of Veronica elliptica is quite distinct from any with which I am acquainted, it having remarkably large leaves. As it does not appear to vary to any extent, and since the climatic conditions of the Open Bay Islands are not such as might be expected to favour a luxuriant development of leaf, but rather the contrary, most probably this is a constant variety and would “come true” from seed. It would be consequently a matter of considerable interest to procure seeds and raise plants, so as to settle this question. Also, plants could readily be propagated from cuttings, and then grown under different conditions, so as to observe if the leaf-form could be easily modified. Asplenium obtusatum grows densely in many places beneath the bushes of Veronica associated

with Astelia nervosa.* Syn. Astelia grandis, Hook. f. Frequently, too, Pteris incisa is mixed with the Asplenium, above the dark-green fronds of which its much paler green leaves rise to a height of 20 cm. or more. Very frequently the Veronica is accompanied with, or indeed it may be entirely replaced by, Muehlenbeckia adpressa associated with Pteris incisa and Asplenium obtusatum, which the liane binds together into a dense thicket of greenery, above which its ultimate shoots are raised for some 28 cm., at first straight, but finally winding towards their apices. Such a thicket, especially if Veronica be present, is about as tall as a man of medium height. (c.) Liane Formation of Freycinetia banksii. By far the greater part of the larger island is occupied by a most interesting formation of which the liane Freycinetia banksii is the dominant plant, and it is the existence of this and the above-described Muehlenbeckia formation that makes the Open Bay Islands of special phytogeographical interest. This Freycinetia formation forms a “scrub,” ± 1.9 m. tall, of the most extreme density, the stout shoots, ± 12 cm. in circumference, twisted in all directions. Sometimes there is little else than Freycinetia, but frequently the scrub becomes still denser through an admixture of Muehlenbeckia adpressa and Calystegia tuguriorum. Between the Freycinetia formation and the cliffs or shore is frequently “Veronica scrub,” or in places Pteris incisa, with Phormium tenax growing through it. On the island are also a few plants of Cordyline australis, and one plant of Schefflera digitata was noted. I think also that Mr. Roberts told me he had noticed a plant of Metrosideros lucida when making a survey of the island many years ago. 3. Vegetation of the Smaller Island. The most interesting fact regarding the vegetation of the smaller island is that although the principal plant-society is a liane formation, yet it is Muehlenbeckia and not Freycinetia that is dominant, the latter being, so far as I saw, altogether absent, just as it is on the smaller half of the larger island. The island under consideration, as pointed out above, consists of a conical hill, and at low water flat rocks extend uncovered for some distance seawards. On these rocks, but just above high-water mark, is Lepidium oleraceum associated with large masses of Tillæa moschata. Asplenium obtusatum is also on these flat rocks, and in some places near the sea is Carex comans.

In one place where the ground is boggy are large quantities of Carex ternaria* This also occurs in some low ground on the larger island in the midst of “Muehlenbeckia-Veronica-Pteris scrub.” together with Hydrocotyle americana,† I am indebted to my friend Mr. D. Petrie, M.A., for the identification of this species. but where the ground is drier at this point is Muehlenbeckia complexa growing over Lepidium oleraceum and Asplenium obtusatum. Excepting in the immediate neighbourhood of the shore the remainder of the island is occupied by a dense growth of “Muehlenbeckia scrub,” at first mixed with Veronica elliptica, but afterwards, towards and at the actual summit of the island, it is almost pure Muehlenbeckia. 4. Origin of the Liane Formations.‡ A preliminary account of this was given in a paper read by me at the Dunedin meeting of the Australasian Association for the Advancement of Science, and published in the Canterbury Agricultural and Pastoral Association Journal, vol. vi., 1904, p. 7. That the Open Bay Islands have formed a part of the South Island mainland at no very distant date, as stated above, there can be no doubt: the shallowness of the sea, the glacial deposit on the main island, and other facts connected with the general geological history of New Zealand, which need not be cited here, offer ample proof. The islands when forming part of the mainland would be densely covered with an evergreen subtropical forest, as mentioned before, similar to that which occupies the whole coast of south Westland at the present time, and which is very briefly described at the beginning of this paper. In such a forest, it may again be pointed out, Freycinetia banksii is a most conspicuous feature, climbing the highest trees and in some places spreading over the forest-floor. Muehlenbeckia adpressa, too, as in nearly all New Zealand lowland forests, is by no means uncommon. With the separation of the Open Bay Islands from the mainland the forest would be at first exposed to few deleterious influences, but as the area that it occupied became gradually smaller and smaller, so would the forest be exposed more and more to the furious north-west wind laden with seaspray, until, by slow degrees, those plants unable to endure such conditions, such as most rain-forest hygrophytes, would go to the wall, while those plants possessing a more xerophytic structure would be better able to survive. Notwithstanding that lianes are amongst the most highly specialised of forest plants, their whole organization having been evolved by forest conditions, they are usually more xerophytic than the trees

with which they are associated,* Thus Warming writes: “Blattbau und Sprossbau eines Teiles der Lianen erinnern an den Bau der Xerophyten.” Lehrbuch der ökolo-gischen pflanzengeographie, Zweite Auflage der Deutschen Ausgabe, 1902, p. 110. since by virtue of their comparatively slender but extremely long stems they have to possess special adaptations in order to procure a sufficient supply of water for their leafy shoots high on the tree-tops. Thus, as the forest trees gradually died, the lianes in question would become dominant, and, uniting themselves with certain coastal plants, and evading the wind by no longer climbing up aloft, be able to maintain themselves in comparative luxuriance. Leaving out of the question any xerophytic structure, this power of certain lianes to assume a shrubby habit, through the interlacing or twining of their shoots, gives them a great advantage over forest trees when exposed to furious gales. Further evidence of the ancient forest formation is afforded by the plants of Schefflera digitata and Metrosideros lucida. Phormium tenax, Cordyline australis, and Arundo conspicua would, on the contrary, be rather plants of the forest's outskirts, Phormium tenax especially occurring frequently as a belt near high-water mark. Judging from the ordinary coastal plant-formations of south Westland, there should appear nothing surprising in the Freycinetia formation of the Open Bay Islands, insomuch as Freycinetia in many places forms an outlier of the present forest formation, in company with certain shrubs and low trees, growing very close indeed to the sea. Muehlenbeckia adpressa, too, frequently remains as a somewhat ball-shaped bush of the open where the forest has been cleared, and in some places it too comes to high-water mark, as e.g. near Nugget Point, in Otago, where, with the shrubby nettle, Urtica ferox, and certain ferns, it forms dense thickets, probably in places where the forest has been removed. Muehlenbeckia complexa, too, on the Port Hills, Canterbury, and elsewhere, no longer a forest plant, forms dark-coloured round or pyramidal shrubs, which are so frequent on the tussock slopes as to give a peculiar physiognomy to the landscape. Many other examples of lianes taking a shrub form in the open after the removal of forest could be adduced, but the above must suffice here. An interesting fact in the distribution of plants on the Open Bay Islands, and one for which I can offer no adequate explanation, is the non-occurrence of Freycinetia on the smaller island. On the larger island, also, it does not occur on the smaller part. It may be that in the struggle for existence the Muehlenbeckia may have some slight advantage over the Freycinetia which

we are unable to estimate and that where the two plants are associated the latter is being slowly eradicated. Such silent conflicts, of which we know nothing, must be constantly waged between plants which apparently are equally well equipped for the struggle. 5. Summary. (1.) The Open Bay Islands are two small islets three nautical miles from the coast of south Westland, to which they must have been joined at no very distant date. (2.) Their most important vegetation consists of thickets formed by the lianes Muehlenbeckia adpressa and Freycinetia banksii. (3.) On the larger island Freycinetia is dominant, associated with Muehl. adpressc, Calystegia tuguriorum, and Pteris incisa. or it may be almost pure. (4.) On the smaller island there is no Freycinetia, but the thicket consists of Muehlenbeckia adpressa associated with Veronica elliptica and Pteris incisa, or the Muehlenbeckia may be pure. (5.) The form of Veronica elliptica is distinct from any found elsewhere in the New Zealand biological region. Probably this form may reproduce itself “true” from seed. (6.) Owing to the heavy rainfall there is in places a luxurian rock-vegetation. (7.) In certain places not occupied by “liane scrub” there is a well-marked zonal distribution of certain New Zealand coastal plants. (8.) When attached to the mainland the present islands must have been occupied by subtropical evergreen rain-forest similar to that now existing on the adjacent coast. After separation, as the area of the islands became smaller and smaller, and the climatic conditions more and more severe, only those plants specially adapted to such conditions could survive, and of these certain of the lianes, although most highly specialised forest plants, are the most suitable. (9.) On the smaller island the dominant liane is Muehlenbeckia adpressa; and Freycinetia banksii, the dominant liane of the larger island, is altogether wanting. 6. List of Plants (Spermaphyta and Pteridophyta) Observed on the Open Bay Islands. Filices. Asplenium obtusatum. Forst. Lomaria dura, Moore. Pteris incisa, Thunb.

Pandanaceæ. Freycinetia banksii, A. Cunn. Gramineæ. Poa annua, L. (adv.). Agrostis (?) sp. Arundo conspicua, Forst. †. Cyperaceæ. Scirpus nodosus (Br.), Rottb. Carex comans, Bergg. Carex teranaria, Forst. †. Liliaceæ. Cordyline australis (Forst.), Hook. †. Astelia nervosa, Banks and Sol Polygonaceæ. Muehlenbeckia adpressa, Lab. Muehlenbeckia complexa (A. Cunn.), Meissn. Rumex flexuosus, Soland. Cruciferæ. Cardamine hirsuta, L. Lepidium oleraceum, Forst. †. Crassulaceæ. Tillæa moschata (Forst.), DC. Myrtaceæ. Metrosideros lucida (Forst.†), A. Rich. Araliaceæ. Schefflera digitata, Forst. Umbelliferæ. Apium prostratum, Labill. Apium filiforme. Hook. Primulaceæ. Samolus repens (Forst.), Pers. Convolvulaceæ. Calystegia tuguriorum (Forst.), R. Br.

Scrophularinaceæ. Veronica elliptica, Forst. †., var. Compositæ. Sonchus asper, Hill. Explanation of Plate XXIII. Vegetation near edge of cliffs of larger island, showing zonal arrangement of plants, Open Bay Islands In centre, the fern Asplenium obtusatum; in foreground and to left, the sedge Carex comans; nearer to edge here and there may be seen the zone of Tillæa moschata; in background, shrubs of Veronica elliptica, var., as noted in paper.