The Vegetation of New Zealand
Chapter I. — General Remarks
The extremely mountainous character of New Zealand has led to the development of a rich and varied high-mountain flora, no fewer than 511 species never descending so far as records go — into page 225the lower lowland-belt, while, in addition, 115 species occur in the lowlands only under special circumstances. There are also the 339 species, partly dealt with in Section II, Chapter I, which are common to the high-mountain and lowland belts, so that the high-mountain flora as a whole numbers 965 species as against 1058 for the lowlands and lower hills, but if the number of species belonging to the high-mountain and coastal elements of the latter be deducted, the high-mountain flora exceeds that of the lowland-lower hills by 68 species.
If the 511 purely high-mountain species be alone considered, they belong to only 38 families and 87 genera, but if the 115 occasional lowland but really high-mountain species be added, the number of families and genera rise to 47 and 114 respectively. No less than 495 of the 511 (nearly 95%) species are endemic; but the endemism of many provisionally non-endemics is highly probable, e. g. Scirpus crassiusculus, Epilobium tasmanicum (a considerable mixture), Myosotis australis (another mixture), Plantago Brownii (still another mixture) and Craspedia alpina — the genus highly polymorphic with its many jordanons and hybrid swarms. In addition to the species, at least 120 groups of hybrids are present, taking the whole flora into consideration, and many are extremely common, while no less than 42 genera are concerned.
The following genera are confined to the high-mountain belt (endemic marked*): — Marsippospermum, Exocarpus, Hectorella*, Pachycladon* Notothlaspi*, Corallospartium*, Swainsona, Pernettya, Mitrasacme, Logania, Pygmaea*, Phyllachne, Haastia*, Leucogenes*, Ewartia and Traversia*. This list is quite trifling when compared with that of the lowland-lower hills with its 23 families and 107 genera, but the following two lists bring out more clearly the striking floristic differences between the two floras, and, at the same time, with what has gone before, show the special composition of the high-mountain flora.
1. Genera typical high-mountain which descend to sea-level or there-abouts under special ecological conditions, particularly a semi-subantarctic climate. Alsophila, Cystopteris, Triodia, Carpha, Oreobolus, Gaimardia, Lyperanthus, Adenochilus, Claytonia, Caltha, Drapetes, Actinotus, Penta-chondra, Archeria, Liparophyllum, Donatia, Ourisia, Abrotanella and Crepis.
2. Characteristic high-mountain genera with the number of species for the genus in brackets followed by the number confined respectively to the high-mountain and lowland belts. (Coastal species are included with lowland, but endemics of the outlying islands are excluded) Isoetes (2), 1, 1; Agrostis (8), 5, 1; Deyeuxia (11), 4, 3; Deschampsia (5), 2, 0; Trisetum (5), 3, 1; Danthonia (15), 7, 1; Triodia (4), 2, 0; Poa (25), 8, 6; Agropyron (4) 2, 1; Uncinia (19) 8, 6; Carex (52), 19, 16; Luzula (13), 9, 1; Colobanthus (10), 6, 2; Ranunculus (45), 30, 14; Nasturtium (most likely in part an unnamed endemic genus) (7), 5, 2; Geum (6), 4, 1; Acaena (12), 5, 1; Pimelea (16), 8, 4; page 226Drapetes (4), 3, 1; Epilobium (38), 14, 9; Schizeilema (11), 8, 1; Aciphylla(28), 26, 0; Anisotome (17), 12, 4; Cyathodes (4), 2, 0; Dracophyllum (23), 12, 2; Gentiana (19), 14, 1; Myosotis (32), 24, 4; Hebe (67 as defined at present but there are certainly more), 48, 14; Veronica (12), 7, 2; Ourisia (12), 8, 1; Euphrasia (13), 11, 2: Plantago (8), 3,1; Lobelia (3), 2, 1; Forstera (4), 3, 0; Olearia (37), 12, 15; Celmisia (56), 39, 3; Raoulia (23), 17, 1; Helichrysum (11), 6, 2; Cotata (19), 6, 8; Abrotanella (7), 5, 1; Senecio (32), 9, 12.
Certain other genera, or families, though they contain few or no high-mountain species play an important part in the vegetation, e. g. Podocarp-aceae, Rubiaceae, Festuca, Phormium, Chrysobactron*, Herpolirion, Notho-fagus, Elytranthe, Muehlenbeckia, Stellaria, Drosera, Carmichaelia, Discaria, Geranium, Oxalis, Coriaria, Stackhousia, Aristotelia, Viola, Hymenanthera, Leptospermum, Halorrhagis, Nothopanax, Hydrocotyle, Angelica, Corokia, Griselinia, Gaultheria, Leucopogon, Cyathodes, Epacris, Archeria, Suttonia, Pratia, Wahlenbergia, Lagenophora, Brachycome,, Gnaphalium, Cassinia, Craspedia, Microseris and Taraxacum.
The headquarters of the true high-mountain flora of 511 species is in the lofty ranges of South Island, with their 472 species 379, of which (80%) are confined thereto. North Island possesses only 105. species, of which 29 are not in South Island, and Stewart Island 39, of which 9 are confined to that island. The comparative poverty of the North Island high-mountain flora is most likely owing to the small area suitable for occupation as compared with South Island, and the far lower average height of the mountains is another factor concerned. In South Island, each botanical district shows a good deal of local endemism and the area of distribution of many species is small, so that there is a gradual change in the flora in proceeding from north to south. The greatest differences, both ecological and floristical, are in relation to rainfall, there being what may be conveniently styled wet and dry mountains.
Vertical distribution (the belts of vegetation).
Details regarding vertical distribution are not easy to supply. All-important is the average winter snow-line above which the ground for some months is covered continuously with snow. Below this line the covering is not continuous, though at intervals more or less snow lies on the ground throughout the winter. The winter snow-line, and the average periods which snow lies at various altitudes below that line, are correlated with the aspect of the slope, so that alpine species descend much lower on shaded than on sunny faces. Edaphic conditions, hollows, close proximity to shrubs or rocks &c. also contribute their share, e. g. bogs, shingly river-beds and rocks may bring high-mountain species, and even associations, into the montane or even the upper lowland belt (see under next heading). Even on forest-clad mountains many bare patches, containing high-mountain grass or herb vegetation, extend far below the forest-line. Then there is the gradual effect of change page 227in latitude and the differences in distribution on wet or dry mountains, as also the many intermediate stages between such extremes. In fact, each mountain supplies its own special circumstances, and, were the details at my disposal far more accurate, only general statements could be made. As it is, one has to trust in many instances to estimated heights, so that the details given here and further on are essentially approximate.
Commencing at sea-level the belts of vegetation are here styled, lowland, montane, lower subalpine, upper subalpine and alpine. Obviously, the boundaries of these differ for each botanical district and also in the districts themselves, as also on different parts of the same mountain, so that no actual altitudinal limits can be defined. The chief delimiting factor appears to be the average length of time the winter snow lies upon the ground, and in the upper alpine belt of the highest mountains this covering persists far into the summer. This reliance on the different snow-lines as a basis for the delimitation of the belts in supported by the fact that as soon as that part of a mountain is reached certain characteristic species are encountered which are rare or absent at a lower altitude1, and also a new vegetation is met with; so, too, though in a somewhat lesser degree, for distribution in regard to the other winter-snow limits.
The average line on any mountain, but different for different aspects, at which the maximum occupation of the ground by winter snow begins marks the commencement of the true alpine belt, and this extends either to the perpetual snowfields ore the summit of the mountain. Below this line comes a second one which denotes that there has been a covering of snow for a lesser period than on the alpine belt and this line marks the commencement of the upper subalpine-belt. Below the latter, there is usually no continuous covering of snow for more than a week or two at a time, and the average line marking this area is the commencement of the lower subalpine-belt. Below the latter is an area where snow lies only on an average for a few days at a time the average lowest line of which marks the commencement of the montane belt, below which lies the lowland belt, in the upper part of which snow lies on an average for a day or two at most, but in the north of North Island, and on its coast snow never falls in this belt.
1 1) For instance in the North-western district, Danthonia australis, which even on mountains to the east may fill hollows where snow lies for a long period; in the Eastern district Celmisia Haastii and C. viscosa; in the Western and parts of Fiord districts, Danthonia crassiuscula and Celmisia sessiliflora and in the Fiord district Ranunculus Buchanani, R, Simpsonii and Celmisia Hectori.
Taking the whole high-mountain flora of 1058 species an estimate for the number of species in each belt is as follows: — lower-subalpine 725 (doubtless certain forest species included here belong rather to the montane belt); upper-subalpine 496; alpine 251, of which perhaps 100 occur in its uppermost part. The lowland-high mountain element plays only a comparatively small part in the upper subalpine and alpine belts with its 130 species (59 are virtually high-mountain) in the former belt and 27 species (15 virtually high-mountain) in the latter. Taking the purely high-mountain species alone, 178 species occur in the montane belt, 361 in the lower subalpine, 366 in the upper subalpine, and 224 in the alpine.
High-mountain plants at sea-level.
As already seen, 115 species of high-mountain plants occur at about sea-level, many most characteristic alpine and subalpine species, e. g. to mention a few, Carpha alpina, Oreobolus pectinatus, Gaimardia ciliata, Astelia linearis, Caltha novae-zelandlae, Carmichaelia Monroi, Drapetes Dieffenbachii, Cyathodes empetrifolia, Coprosma repens, Donatia novae-zelandiae, Celmisia argentea and Senecio Lyallii. There is also a second category, the members of which are nearly or quite numerous enough to class with lowland species, although their distribution may be restricted to a special climate or soil. Such are, Podocarpus nivalis, Dacrydium Bidwillii, Phyllocladus alpinus, Astelia Cockaynei, Carmichaelia grandiflora, Coriaria lurida, Hoheria glabrata, Pimelia Gnidia, Gunnera dentata, Gentiana Townsoni, Hebe Raoulii, Olearia Colensoi and river-bed species of Raoulia. These lowland-high-moutain plants fall into three principal classes as follows: — (1.) Those which as plants of tussock-grassland find a continuous path by means of that formation to the lowlands. (2.) Those which descend by means of stony river-bed. (3.) Those which are restricted to that part of New Zealand possessing a modified subantarctic climate. To the first two classes belong xerophytes of physically dry stations, and to the third class shrubs of the subalpine-scrub and so-called "bog-xerophytes". Actual high-mountain associations, and not isolated species only, occur at sea-level, and such have already been briefly described in Section II Chapter IV, and veritable high-mountain species, or their equivalents, are present on coastal cliffs (Section I, Chapter IV). As for the causes furthering the presence of high-mountain species at low altitudes something is said under the next head.
The ecological conditions of the high mountains.
High mountains the world over are subject to a set of similar and fairly definite conditions which there is no need to discuss here, especially as they are so admirably set forth by Schroter in his splendid work Das Planzenleben der Alpen, 1925–26: 920–1027. On the other hand, those specially affecting New Zealand need brief mention, so far as they are known.
No accurate details are available regarding the high-mountain climate. It is clear however that the species are not attuned to nearly so great intensity of cold as are alpine plants in general. This is clearly brought home by the fact that many species of the alpine belt cannot endure the winter temperature of Kew and hardly any that of Berlin. Probably — 18° C. is more than most can endure. For example, the winter of 1923 did more damage than usual in the neighbourhood of Queenstown (SO., lat. 45°, 310 m. altitude). Though there was almost constant frost for six weeks, and many supposedly hardy plants were killed, so far as can be ascertained the shade temperature did not fall below — 11° C. and this statement is strongly supported by the fact that the exotic Eucalyptus Gunnii, juvenile E. globulus (adult trees killed or greatly damaged), and Pinus radiata were undamaged. On the other hand, various species1 which are common above the forest-line were killed or badly damaged, as were the purely lowland and coastal species. Certainly, in the high mountains, a covering of snow stands for a good deal, but many species2 tolerate equally both snowy and snowless growing-places, while many of the less hardy shrubs may be only partly buried.
1 1) Gaultheria perplexa, Leptospermum scoparium, Weinmannia racemose, Notho-panax Colensoi, Shawia paniculata, Phormium Colensoi, Senecio cassinioides (high-mountain only) and Senecio elaeagnifolius. All the species mentioned in connection with this frost were either cultivated plants in the Queenstown Gardens or growing wild in the immediate neighbourhood.
2 2) Ranunculus Grahami, the rosette species of Nasturtium (generic name provisional), Epilobium. rubro-marginatum, several dwarf Aciphyllae, Myosotis macrantha, Hebe pinguifolia, H. epacridea, H. Haastii, cushion species of Raoulia, Leucogenes grandiceps and many others.
The north-west downpour on which depends the distribution of forest on the west of South Island is, until it has crossed the Divide, a warm, snow-melting rain, whereas that from the south-west is cold and generally terminates in snow, succeeded by frost. This latter rain, in South Otago and Stewart Island is a most important factor towards inducing herb-ficiu and bog. Its frequent occurrence, combined with cloudy skies, brings in those subantarctic conditions which even in the lowlands favour high-mountain species, indeed it is not going too far to declare that where the climate is of this character the occurrence of such species is rather a matter of opportunity than of altitude, and given ground to grow upon they will become established and form associations almost anywhere (L. Cockayne, 1925: 79, 80). On the slopes of the Southern Alps facing the Canterbury Plain and on the Seaward Kaikoura Mts. much rain comes from the east.
The wind-factor is of great moment, acting as it does both mechanically and physiologically upon the plants, while as an agent of denudation it is of considerable importance. Hemmed in between the high ranges, bounding narrow valleys, the power of the wind attains great intensity. Bare ground is frequently impossible to populate with plants when swept at intervals by furious gales. On exposed ridges and low mountain tops wind forbids a close covering and allows the establishment of dwarf species of plants of the prostrate, cushion and creeping forms, which belong more properly to the alpine belt. The north-west wind of South Island is of peculiar ecological significance. But for its prevalence, there would be forest where tussock-grassland at present rules. When this wind, a true foehn, rages, and the sky is cloudless, under a burning sun transpiration must reach its maximum. So violent is this wind that it is impossible to stand upright on a ridge exposed to its full blast. Even in an extreme forest climate where the wind strikes with its maximum power tussock-grassland may be established yet with forest on either side where the force of the wind is somewhat less (Fig. 49).
Snow is a most important factor in New Zealand high mountains. It has been already shown how the primary division of the vegetation into belts depends upon the average duration of the snow-covering. When the alpine belt is almost bare of snow at the end of January or much later in the Western, Fiord and South Otago districts deep masses of snow, at times resembling glaciers, still lie in gullies and cirques. Haast (1886: 30) found by actual measurement 12 m. depth of snow on one of the passes of the Southern Alps, and that in a gorge he estimated as being 150 m. deep. page 231Snow avalanches are extremely frequent, even on the driest mountains, and their effect is great both in destroying vegetation (Fig. 50) and furthering denudation while, as agents of distribution, they bring even living tussocks and shrubs into the valleys where they occasionally become established. In the upper subalpine and alpine belts, the vegetation after the snow has melted looks just as if a steam-roller had passed over the surface, flattened to so the ground are the plants, including tussocks of Danthonia Raoulii var. flavescens 1.2 m. high. The effect of temporary streams and pools or of snowwater in hollows, is reflected by the presence of special species and combinations of plants. The mechanical effect of a heavy snow-covering upon the subalpine-scrub may be pointed out, but it alone is not responsible for the peculiar life-forms.
The plants themselves once established modify both climate and soil so greatly as to make their own conditions. For example, though in the first instance, soil be the same and climate the same, the ecological circumstances of fell-field, herb-field, forest and scrub are quite dissimilar. The burning of Nothofagus forest or subalpine-scrub — a natural circumstance if caused by a volcanic eruption — may lead respectively to replacement by tussock-grassland or by a colony of Phormium Colensoi — communities ecologically distinct from the replaced associations. The subantarctic characteristic of a plant's dead parts turning into peat while still attached to the living plant itself is strongly developed in many alpine genera (Phyllachne, Donatia, Celmisia, Raoulia &c.) and plays an important part in modifying the habitat.
Aspect is a matter of fundamental importance, leading, as it does, to local climates in close proximity. This is seen on all sunny and shady slopes (the "sunny" and "dark faces" of the shepherds), but it appears at its maximum in narrow mountain valleys where for more than three months yearly the shady side may receive no direct sunshine and the ground remain frozen hard for many days at a time while the opposite slope is quite warm.
Topographical changes are of course of prime importance both with regard to the evolution of vegetation and perhaps of species. At the present time, the great prevalence of mountains composed of greywacke and allied rocks which supply debris in enormous quantity leads to the constant establishment of migratory communities both progressive and retrogressive. Although such are from their nature transient, yet they are constantly being re-established, so that they present permanent habitats where habitat-effects can accumulate and new species arise, if epharmonic response be an agent in evolution. Disintegration of the surface-soil, the result of wind, snow, frost and rain action, is always in progress, especially where tussock-grassland conditions prevail, to that the surface is not even like a meadow but there are low raised mounds of vegetation surrounded by sunken bare patches. Many other details as to habitat-ecology are cited when dealing with the communities.
Repeopling the new ground during the retreat of the glaciers.
Even today, glaciated New Zealand, at a very low level indeed, is not a thing of the past. Probably, a fair idea of what many valleys of glaciated Westland were like during the retreat of the ice from the coastal plain and lower mountain slopes is afforded by the Franz Josef and Fox glaciers and their immediate neighbourhood with their terminal faces at 211 m. and 204 m. respectively and distant only a few kilometres from the sea. At the present time, the peopling of the new ground just abandoned by the ice can be observed, together with what has taken place at no distant date, indeed every transition can be plainly seen from bare rock, or moraine a year or so old, to forest. It seems then not unreasonable to conclude that what is happening at the present time is merely a repetition of what occurred throughout the Western district at the conclusion of the New Zealand ice-age.
In the case of the Franz Josef glacier — studied by me in 1910 and 1911 — three habitats are being invaded, namely rock smoothed by the ice, moraine, (both lateral and terminal) and river-bed, the rock being by far the most extensive. At an altitude of about 300 m. close to the abandoned rock (quartzose schist marked by numerous cracks, grooves and notches running parallel to the ice) there is no vegetation, but at a few metres distance from the ice, there are everywhere patches large and small some 2 to 3 m. deep of the moss Rhacomitrium symphiodon. This plant clings to the rock with great tenacity; its leaves when wet are spreading and hold much water in their axils, but when dry, they are erect and pressed closely to the stem. When the moss, through its rapid decay has prepared a seed-bed, the chinks in the rock are invaded by vascular plants (Fig. 51), the "seeds" brought by wind from the neighbouring scrub and forest or carried on the rock-surface by water. More than 30 species of pteridophytes and sper-mophytes take part in the invasion1.
1 1) The following are the most important: — Eymenofhyllum multifidum (grows on solid rock and Forms soil), Lycopodium varium, Deyeuxia pilosa, Poa novae-zelandiae, P. Cockayniana (forms large mats), Schoenus pauciflorus (especially where water lies), Earina autumnale, Carmichaelia grandiflom, Coriaria arborea, Metrosideros lucida, Gunner a albocarpa (broad, rooting mats), Gaultheria rupestris, Dracophyttum longifolium, Hebe subalpina, Veronica Lyallii, Coprosma rugosa, Celmisia bellidioides, Olearia avicenmaefolia (almost the first to arrive), 0. ilicifolia, O. arborescens and 0. Colensoi. All these species occur in the immediate neighbourhood of the glacier and, except the arina, belong to the subalpine florula. No species, not belonging to the locality has been observed.
On older moraine-covered rock, at some distance back from the rock now being invaded, is a broad belt of tall scrub consisting of the shrubs already mentioned (subalpine-scrub species), together with rain-forest species, especially Asplenium bulbiferum, Blechnum lanceolatum, Polystichum vestitum, Histiopteris incisa, Carpodetus serratus, Weinmannia racemosa, Melicytus ramiflorus, Fuchsia excorticata, Coprosma lucida and Coprosma foetidissima. Within, is more or less Metrosideros lucida, i. e. the association is potential southern-rata forest. Such an association forms the next belt which extends upwards perhaps to the scrub-line and marks a comparatively recent advance of the ice.
According to Bell (1910:5) "probably not more than 150 years ago", the glacier extended 820 m. northwards down the valley, depositing on its retreat extensive terminal moraines. On these, and the old river-bed, can be seen vegetation at different stages of formation, the climax, so far, being a scrub about 3.6 m. high (Fig. 52). On older moraine still, there is Metrosideros lucida forest but the climax-association of the valley is dicotylous-podocarp forest with Dacrydium cupressinum, Podocarpus ferrugineus, P. Hallii and the ordinary trees, shrubs, tree ferns (Hemitelia Smithii)1 and ferns of Western district rain-forest proper.
1 1) Hemitelia Smithii also occurs in the forest above the glacier.
2 2) The presence of a tree-fern, Hemitelia Smithii, in one valley of the Lord Auckland Islands, is a pretty sure sign, as Speight has pointed out to me, that the ice-period of that group was not due to increased cold.
On the east of the Divide, occupation by plants of the morainic matter of valley floors can be readily observed at considerably less than 900 m. altitude near the terminal faces of certain large glaciers. In such a habitat-complex, many species can be established as colonists including both those of the high mountains and those of the montane belt. Also, as already described, the process of occupation of river-bed, fans, gravel-plains, and even clay hillsides, can be daily seen in operation, and the important role noted which is played by mat-plants and cushion-plants in providing seed-beds.
1 1) Most likely, as now near the Franz Josef glacier, there would be low hills and slopes which would be clad with pre-glacial forest and from this seeds would be carried to the new ground. If this be so, the present podocarp-dicotylous forest of lowland Westland is more or less representative of the first post-glacial forest, and has not replaced a previous Nothofagus succession.