Go to National Library of New Zealand Te Puna Mātauranga o Aotearoa
Volume 48, 1915
This text is also available in PDF
(2 MB) Opens in new window
– 166 –

Art. XIX.—Notes from the Canterbury College Mountain Biological Station.

No. 4.—The Principal Plant Associations in the Immediate Vicinity of the Station.

[Read before the Philosophical Institute of Canterbury, 1st December, 1915] Plates XV, XVI.

(A.) General.

In his general account of the Canterbury College Mountain Biological Station, Dr. C. Chilton points out—Trans N. Z. Inst, vol. 47 (1915), p. 333—that, when the boundaries of the botanical reserve are defined and surveyed a botanical map of the area will be prepared. Such a map would, of course, be based on the distribution of the florula as determined by that of the plant associations. A classification and preliminary account of these associations is, then, a necessary prelude to that detailed study which the preparation of a map of the vegetation demands. The associations as defined below, although the outcome, in part, of the great topographical changes which have befallen the area during and subsequent to its period of intense glaciation, are considered from the static and not from the dynamic standpoint, since for purposes of graphic representation they are distinct vegetation entities for the time being. This viewpoint does not in the least preclude the ultimate necessary study of the plant-covering with regard to its evolution; it merely offers a convenient basis for present investigations No attempt is made at thoroughness in our treatment of the associations, which is descriptive merely; nevertheless we believe that brief, incomplete descriptions of undescribed plant communities are better than nothing, and can be of considerable assistance to those engaged in comparative studies of New Zealand vegetation. Each association is a collection of species bound to a definite habitat, the latter being the sum total of the various ecological factors to which the association is exposed. “Habitat,” as thus defined, is virtually impossible to determine accurately, but it is roughly measured by climate, by the nature of the soil in its widest sense, by the relation of plant to plant, and by the animals present. But, even were we able to estimate the exact results of habitat, no such great advance would be made as might be supposed, since each species exists under a distinct environment of its own. Such individual environmental differences may, in fact, easily be far greater within one association than are the general habitat-influences of two adjacent distinct associations.

The Canterbury College Mountain Biological Station is situated in the Eastern South Island Botanical District,* close to its junction with the

[Footnote] * The botanical districts referred to in this paper may be provisionally defined as follows: (1.) The North-eastern South Island Botanical District includes the northeastern portion of the South Island, excepting the wet area in the vicinity of the Marlborough Sounds, and it is bounded on the west by a line marking the average limit of the western rainfall, and on the south by the River Waiau. (2.) The Eastern South Island Botanical District is that area extending from the Waiau to the Waitaki Rivers, and terminating on the west at the line reached by the average western rainfall. (3.) The Western South Island Botanical District extends from the River Taramakau in the north to a line not yet determined, lying somewhat to the south of the River Haast, and on the east it crosses the actual divide and is bounded by a line marking the average limit of the western rainfall —L. C.

– 167 –

Western South Island Botanical District, or, in other words, it lies just beyond the influence of the excessive western rainfall. This is most plainly reflected in the physiognomy of the vegetation, where, on the Cass side of the River Waimakariri, the xerophytic grass-tussock form dominates the landscape (see Plate XV, Fig. 1); while on the other side of the river, only some two miles distant, stands a vast dark mesophytic forest-mass with clean-cut margin, hardly one tree standing out into the open grassland of the east. Nowhere can a more striking example be seen of the direct effect of climate. The change in physiognomy is instantaneous; there is no transitional phase.

The matter of this paper is the outcome of a few visits to the area made at different times by the authors independently of one another, and especially of a recent brief stay at the station, when the various associations were conjointly examined by the authors, and lists of the species hurriedly taken. Unfortunately, the time available for observations was much too short, while, owing to the season of the year (November), certain species were not in bloom and could not be accurately determined. Therefore, this study is limited to brief descriptions of the associations, and doubtless many species have been omitted. In certain cases “var.” is placed after the specific name. This denotes that the species is an aggregate, and that the plant referred to is one of the unnamed varieties of which the aggregate is composed. On the other hand, “var.” being absent does not mean definitely that the species is a true entity. It means either that we think the species may be a definite entity or else that we have no proof that it is an aggregate.

The associations dealt with fall under the following heads: (1.) Forest, (2) shrubland, (3) grassland, (4) swamp, (5) rock, (6) river-bed. This classification is obviously faulty, since it contains associations based some on growth-forms and some on habitat; but it is convenient, easy of application, and can be readily used both by those visiting the station and for phytogeographical purposes generally.

Many of the associations are greatly modified at the present time through the action of man, the chief modifying agencies being the grazing of sheep and frequent fires. To a much lesser degree, partly also through the above two factors, various foreign species of plants have gained a firm footing and come into competition with the indigenous element. Much more important, however, is the fact that the degree of frequency of the component species of certain associations has been greatly changed, so that infrequent members of the primitive association may now be of prime importance. But it must be pointed out that, so far as ecological studies at Cass are concerned, the introduced plants are now as much members of the associations as are the indigenous species, and they must be equally considered. This is especially important since a good deal of future research may have a distinct economic bearing. This paper concludes with a list, quite incomplete, of the species belonging to the associations. Also, for the information of those not conversant with the New Zealand flora, a very brief account is given of the growth-form of each species. This, it must be pointed out, refers solely to the plant as it grows at Cass, and has nothing to do with the growth-forms of a species through its entire range, or with its commonest form.

(B) The Associations.

(1.) Forest.

In the immediate vicinity of the biological station forest is poorly represented. This is not on account of its destruction by fire, as in so many parts of New Zealand, but because in the Eastern Botanical District, in the

– 168 –

montane and subalpine belts, the ecological conditions are antagonistic, and forest occurs only in specially well-sheltered situations. Here only the three small pieces of forest situated in the gullies at and near the base of Mount Sugarloaf are dealt with.

The association belongs to the mountain southern-beech (Nothofagus cliffortioides) formation, a plant-community of wide distribution in the high mountains of both the North and South Islands. At Cass the association, so far as its undergrowth is concerned, was probably always far from dense, but now, through sheep having sheltered for many years in the forest, undergrowth in no few places is wanting. But the plant-covering of certain spots gives a fair clue to the primitive physiognomy of the forest interior, with its slender tree-trunks and moss-covered floor occupied by numerous Nothofagus seedlings, small fronds of Blechnum penna marina, and erect or straggling bushes of Coprosma parviflora, C. microcarpa, Aristotelia fruticosa, Clematis marata as a ground-plant, and Rubus subpauperatus.

The sole tree is the mountain southern-beech, Nothofagus cliffortioides. Parasitic on it are the woody hemi-parasites Elytranthe flavida and E. tetrapetala. The forest-floor is usually dry, and consists on the surface of a brown humus held together by numerous beech-rootlets, while beneath is the usual clay of the locality.

On the margins of the streams the vegetation, as might be supposed, is much richer than elsewhere. The following is a list of the species noted near such streams: Blechnum penna marina, B. capense var., Polystichum vestitum, Lycopodium Selago, Carex dissita var., Corysanthes macantha, Urtica incisa, Cardamine heterophylla var., Acaena Sanguisorbae var pusilla, Rubus subpauperatus, R. schmidelioides var. coloratus, Aristotelia fruticosa, Hymenanthera dentata var. alpina, Epilobium rotundifolium, Galium umbrosum, Coprosma parviflora, Veronica salicifolia var communis, V. Traversii var., Erechtites glabrescens.

(2.) Shrubland.

By “shrubland” is here meant those associations where the shrub-form dominates. It differs greatly in its physiognomy and physiological characters according to the degree of closeness at which the component shrubs occur and their growth-forms. Where the shrubs as a whole do not touch one another, but stand as small clumps or single plants dotted about here and there, the association may be termed “open shrubland” Where the shrubs grow closely, but do not form dense entanglements, the term “thicket” may be used; but where there is a dense interlacing mass of branches, or the divaricating growth-form plays a notable part, the suggestive term “scrub” may be applied. These names for the various classes of shrubland, as explained farther on, differ in part from those used by L. Cockayne in his previous writings. Obviously, between all the classes intermediate forms must occur.

Shrubland is generally of a more xerophytic character than forest, but its habitats differ so greatly in this regard that some associations are distinctly mesophytic, while others are strongly xerophytic. Of course, it is impossible without physiological, anatomical, cultural, and ecological investigation to declare with confidence that any plant is either a xerophyte or a mesophyte. But, undoubtedly, when it is seen that examples of one and the same true-breeding entity assume different growth-forms under mesophytic and xerophytic conditions, and that these growth-forms approach

Picture icon

Fig 1—Cass Plain with low tussock grassland, greatly modified by grazing On left, Discana toumatou Raoul Cragiebuin Mountain and gorge of River Cass in background
Fig 2—Prostrate and erect varieties of Podocaipus nivalis Hook growing on stony debris slope at about 3,000 ft altitude on Mount Sugailoaf Low tussock grassland in background

– 169 –

nearer and nearer to growth-forms of species limited to the one environment or the other, then it is safe to conclude that the induced apparently epharmonic* forms are truly xerophytic or mesophytic, as the case may be. Certain species from the Cass shrubland exhibit such induced xerophily or mesophily to an astonishing degree, as L. Cockayne showed some time ago.

A most interesting case of a contrary character came under our notice while studying the Cass shrubland for the purposes of this paper. On coarse stony debris, as will be seen farther on, Podocarpus nivalis, the mountain-totara, is extremely abundant, its shoots far-spreading and closely hugging the ground in espalier fashion. This, so far as we had previously noted, was the invariable behaviour of this shrub in such a station. But, according as the plant grows in sun or shade in various parts of New Zealand, so does it assume the prostrate or the more or less erect habit. However, on this debris slope at Cass there are certainly two distinct hereditary races of the species, one the typical P. nivalis Hook f., which in the above habitat always remains prostrate, and the other, which grows in apparently exactly the same habitat, is a dense shrub 1.8 m. high or more (see Plate XV, Fig. 2). Here, then, a special xerophytic growth-form, the spreading prostrate (espalier), is not, apparently, one whit better suited for its habitat than the more mesophytic, infinitely less wind-resisting, erect shrub-form, which, moreover, from its position should be exposed to a more intense transpiration. The word “apparently” is used advisedly, since the erect shrub may have a longer root-system, and so be under a different environment to the prostrate form.

The Cass shrubland associations are not peculiar to the area, but, with slight floristic modifications, extend throughout the Eastern and Northeastern South Island Botanical Districts, though the latter possesses some shrubland peculiarly its own§ Excepting the open shrubland, and perhaps the Leptospermum thicket, the associations are virtually primitive.

The shrubland associations here dealt with may be designated as follows: (a) Cassinia open shrubland; (b) Discaria (wild-irishman) thicket; (c) Leptospermum (manuka) thicket; (d) river-terrace and debris scrub.

(a.) Cassinia Open Shrubland.

Shrubland where one or other of the closely related species of Cassinia dominate is at the present time a common feature in much of lowland and montane New Zealand, especially on coastal areas from which the forest has been removed by man. In primitive New Zealand, associations

[Footnote] * The term “epharmonic” is here used as in my former writings—e g., “Observations concerning Evolution, derived from Ecological Studies in New Zealand” (Trans. N Z Inst., vol. 44 (1912), pp. 1330)—with a somewhat different significance to that of Vesque and Warming (see “Occology of Plants” (1909), pp. 2 and 369). According to my usage, an epharmonic variation is a change in its form or physiological behaviour beneficial to an organism evoked by the operation of some environmental stimulus. Such a change may be called an epharmonic adaptation, as distinguished from such adaptations as cannot be traced to any direct action of the environment. To the neo-Darwinian no permanent adaptation according to the above definition would be “epharmonic,” whereas to the neo-Lamarckian all would be so considered.—L. C.

[Footnote] † Trans. N.Z. Inst., vol. 32 (1900), p. 123; and ibid., vol. 44 (1912), pp. 15, 17, 20, and 27.

[Footnote] ‡ Trans. N.Z. Inst., vol. 44 (1912), p. 17, and for figure see pl. 4, facing p. 21.

[Footnote] § Especially one on coarse debris where Senecio Monroi, elsewhere a rock-plant, is dominant Another rock species, Helichrysum microphyllum, also occurs in a certain form of subalpine scrub in the same botanical district.

– 170 –

of this character must have been comparatively rare. The species of Cassinia (Compositae) are quick-growing shrubs of ericoid form, tolerant of considerable drought, and bearing in abundance seed of high germinating-power, which can be transported rapidly over considerable distances by the wind.

The association under consideration contains a variety of the coastal Cassinia fulvida as its dominant member. Other shrubby members are Discaria toumatou, Hymenanthera dentata var. alpina, Leptospermum scoparium var., Corokia Cotoneaster, Gaultheria depressa, Styphelia Colensoi, and Dracophyllum uniflorum var.

The association occurs on ground formerly occupied by low tussock grassland, where, through burning the tussocks and subsequent grazing, bare ground suitable for invasion by the Cassinia has been provided The shade and shelter furnished by the last-named encourages the establishment of the other shrubs. The erect shrubby members of the association grow in clumps or as single plants, and the interspaces consist of the original tussock association more or less modified. Some of the species of the latter are: Blechnum penna marina, Poa Colensoi var., Festuca novaezealandiae, Acaena Sanguisorbae var pilosa, Viola Cunninghamii, Pimelea prostrata var., Styphelia Fraseri, Celmisia longifolia var, Raoulia subsericea, Microseris scapigera, and Senecio bellidioides var. glabratus.

Should the Leptospermum become dominant, such an association can be rapidly transformed into manuka thicket, or, on the other hand, with increase of Cassinia, many of the species mentioned above may be suppressed.

(b.) Discaria (Wild-irishman) Thicket.

This is the “Discaria shrub steppe” of Cockayne and Laing,* a suitable enough term, but we think it best to abandon the word “steppe” altogether in New Zealand phytogeography, because there is no consensus of opinion amongst ecologists regarding its usage, so that formations of very different ecological relationships are called “steppe.”

The association consists, so far as shrubs are concerned, almost entirely of the spinous, semi-divaricating Discaria toumatou (see foreground in Plate XV, Fig 1), of an average height of, say, 1.2 m In true thicket the shrubs touch or grow into one another, but frequently this association is more or less open. Clematis marata may be fairly common as a liane Other shrubs of the neighbourhood may be present here and there, but they are of no moment.

The association is extremely constant in its distribution throughout the South Island, except west of the Southern Alps, and, when its dark mass is seen at a distance, indicates dry river-terrace and those stony fans where streams issue from their gorges. Eventually grass tussocks and some of their accompanying plants, especially those of most xerophytic structure, become established in the open spaces between the shrubs.

[Footnote] * Trans. N.Z Inst., vol 43 (1911), p. 349.

[Footnote] † The term “formation,” as used by me, has a much wider signification than “association,” and is used to include such associations as are virtually ecologically similar. Thus the whole rain-forest of New Zealand is one formation made up of many associations which differ from one another floristically, but if ecologically, then only to a very limited extent. So, too, low tussock grassland, swamp, and river-bed are ecological units each containing more than one floristic unit To compare the above phytogeographical conceptions with those of taxonomy, the formation represents the genus and the association the species.—L C

[Footnote] ‡ See E A. Rubel, Journ of Ecology, vol 2 (1915), pp 23335.

– 171 –

(c.) Leptospermum (Manuka) Thicket.

This belongs to the formation hitherto called “manuka heath” by L. Cockayne. It seems to us best to abandon the term “heath,” party because it is of loose application, and partly because in many parts of New Zealand the dominant shrub reaches too great a height to permit comparison with the heaths of Europe. At the same time, where climatic and soil conditions combine to produce a low, more or less uniform growth of L. scoparium, “heath” is far from being an unsuitable designation.

At Cass fairly wide areas occupied by manuka are not uncommon. How far these are primitive we do not know, but some seem to bear a fairly primitive stamp. The relation of Leptospermum scoparium var. to the surrounding vegetation at Cass requires detailed study.

(d.) River-terrace and Debris-Scrub.

On the sides of gullies cut by streams through a fan, on the faces of river-terrace throughout the Waimakariri River basin generally, of which the Cass area forms a part, and on the otherwise bare stony debris of the hillside where weathering has led to an accumulation of rather large stones there are dense scrubs of a similar floristic composition, but differing according to habitat in the relative proportions of the important members. These scrubs owe their special physiognomy largely to the presence of the divaricating growth-form in abundance, and to the green- or it may be glaucous-leaved species of Veronica. Should Veronica dominate—a frequent happening in many parts of montane and subalpine New Zealand—we have a “Veronica scrub.” These scrubs are usually of considerable density. The shrubs grow intermixed, and lianes may bind them still more tightly together. The divaricating shrubs afford an extremely striking example of convergence, belonging, as they do, to several distinct families—e.g., Pittosporum divaricatum (Pittosporaceae), Discaria toumatou (Rhamnaceae), Aristotelea fruticosa (Elaeocarpaceae), Hymenanthera dentata var. alpina (Violaceae), Corokia Cotoneaster (Cornaceae), Coprosma propinqua var. and C. parviflora var. (Rubiaceae), and Olearia virgata var. (Compositae). If this convergence, which we certainly consider epharmonic, were to be dealt with not merely from one plant association, but from the New Zealand region as a whole, the figures would be not merely eight species belonging to seven families, but fifty-six species belonging to nineteen families and twenty-three genera!

The scrub of shaded gullies is closely related to forest, and with the appearance of Nothofagus cliffortioides as a member forest may be considered in process of formation. But the scrub of steep, stony debris, on the other hand, is one of those rather rare communities that is at the same time an initial and climax association on a par with those of shingle-slip, dry rock, and the Epilobium association of river-bed.

According as certain species are dominant, so is the facies of the association under consideration more or less altered. But we do not consider such change great enough to warrant the establishment (except in the case mentioned below) of more than one association for Cass. Besides the shrubs already mentioned, the following are generally present: Discaria toumatou, Gaultheria rupestris var, Veronica salicifolia var. communis, and Veronica Traversi var. Olearia avicenniaefolia var. is occasionally present. If the ground is wet, then Veronica buxifolia var. odora may become so important as to dominate, in which case certainly the association has changed its character, and may be named V. buxifolia scrub.

– 172 –

V. buxifolia, as another variety with erect and but little-branched stems, forms low thicket in wet subalpine western stations, and the two associations must not be confused.

Besides the shrubs there are certain ground-plants in the scrub under consideration. Without giving a special list, the following are characteristic: Cystopteris novae-zealandiae, Blechnum penna marina, Polystichum vestitum, Lycopdium fastigiatum, Acaena Sanguisorbae var. pusilla and var. pilosa, Epilobium pubens, and, if the floor is not too dry, E. chloraefolium.

The lianes of the scrub are: Muehlenbeckia complexa var., Clematis australis, C. marata, Rubus schmidelioides var. coloratus, and R. subpauperatus.

On Mount Sugarloaf there is the interesting open shrub association already referred to growing on a slope of unstable debris, the chief characteristic of which is the dominance of the common prostrate variety of Podocarpus nivalis, forming broad patches, orange-brown in colour, which may grow into one another. The above colour is not permanent, but depends upon the direct action of the sun. Here and there are the erect bushy shrubs of another variety of Podocarpus nivalis, some of which are 1.8 m. high, and 1.2–1.5 m. through (see Plate XV, Fig. 2). Other shrubs present are: Clematis australis prostrate on the stones, Hymenanthera dentata var. alpina, Aristotelia fruticosa, Leptospermum scoparium var., Gaultheria rupestris var., Dracophyllum longifolium (stunted), Styphelia Colensoi, S. Fraseri, Pimelea Traversii, Coprosma propinqua var. Various herbs, &c, occur on the open spaces between the patches of P. nivalis—e.g., Asplenium Richardi, Blechnum penna marina, Lycopodium fastigiatum, Poa sclerophylla (on the finest debris), P. Colensoi var., Epilobium pubens, Myosotis australis var., Celmisia spectabilis, C. Lyallii, Senecio lautus var. montanus. Ecologically the association belongs rather to fell-field than to shrubland, while the finer debris bears true shingle-slip plants.

(3.) Low Tussock Grassland.*

“Tussock grassland” is here used as a substitute for the term “tussock steppe” hitherto used by L. Cockayne in his ecological publications. The reasons for abandoning the word “steppe” have already been given. All the same, when attention is called to the effect of overstocking and burning in increasing the percentage of bare ground to that clothed by tussock, &c., until, as in Central Otago, desert pure and simple is established, there is certainly an induced steppe association in New Zealand, just as there is induced desert. Such induced steppe, however, we would now term “open low tussock grassland.” a term having the merit of defining itself.

The word “low” before “tussock” indicates that the tussocks are either Poa caespitosa or Festuca novae-zealandiae, and not the much taller tussocks Danthonia Raoulii or D. flavescens, which, when dominant, one or the other, make “tall tussock grassland,” a formation absent in the immediate vicinity of Cass.

Low tussock grassland is far and away the most important association in the vicinity of Cass—as, indeed, it is for the whole of the South Island of New Zealand east of the actual divide. Its presence is clearly in harmony with the drier eastern climate, and when the average rainfall and number

[Footnote] * The paper by A. H. Cockayne in this same volume (pp. 15465) should be read in conjunction with this section of our paper, since it opens up and discusses many matters of phytogeographical interest which we have not touched upon.

– 173 –

of rainy days is known for the Cass. Biological Station a fairly accurate measure will be afforded of the maximum toleration of rain by tussock grassland. The description which follows refers chiefly to the association below 3,000 ft. altitude.

The association is at the present time so greatly changed by burning and sheep-grazing that little idea can be formed of its primitive condition, but it seems safe to conclude that the tussocks would generally touch and that Discaria toumatou would be quite as abundant as at present, while Carmichaelia subulata (so readily eaten by sheep), the now rather rare Aciphylla squarrosa, and the grass Agropyron scabrum, then growing through the tussocks, would be plentiful. On the other hand, the extremely abundant small tussock, Poa Colensoi var., and a number of herbs and semi-woody plants, now common, would be far less in evidence, especially Scleranthus biflorus var., Coprosma Petriei (two varieties), and Senecio bellidioides (two varieties or forms).

Theoretically the dominant plant is the tussock Festuca novae-zealandiae, but in reality Poa Colensoi var. is far more plentiful in many places, while over considerable areas the Festuca is absent (see Plate XV, Fig. 1). The following, in addition to species already mentioned as abundant, are common members of the association, the introduced plants being distinguished by an asterisk: Blechnum penna marina (where shady), Lycopodium fastigiatum (where shady), *Holcus lanatus, *Poa pratensis, Carex breviculmis, Luzula sp., Microtis uniflora, Muehlenbeckia axillaris, *Rumex Acetosella, Stellaria gracilenta, Scleranthus biflorus (two vars.), *Cerastium triviale, Ranunculus multiscapus, Acaena Sanguisorbae var. pilosa, Geranium sessiliflorum var. glabrum, Viola Cunninghamii, Pimelea prostrata var., Epilobium elegans? Hydrocotyle novae-zelandiae var. montana, Gaultheria depressa (where shady), Styphelia Fraseri (excessively abundant in many places), Plantago spathulata, Wahlenbergia albomarginata, Celmisia longifolia var, C. spectabilis (where shady), Brachycome Sinclairii, Raoulia subsericea, Gnaphalium Traversii, Cassinia fulvida var. montana, *Hypochaeris radicata (many distinct forms apparently quite apart from change in environment). At its higher levels the subalpine element becomes more abundant, Celmisia spectabilis being especially noticeable, while species confined to shady stations at lower altitudes here grow in the full sunshine.

Where the association occurs on old flood-plain, and the ground is stony, there is often an abundance of Carmichaelia uniflora, and here and there open cushions of C. Monroi. In similar situations, forming a distinct subassociation, thanks to its twitch-like underground creeping stems, the grass Triodia exigua forms a compact even turf. The pretty blue-flowered Veronica pimeleoides var. minor is rather common in stony places, and the easily overlooked Iphigenia novae-zealandiae is probably fairly common The above lists by no means exhaust the florula of the association, but they give a fair idea of its present composition for comparative purposes with other parts of the formation in the South Island, and prepare the way for the necessary detailed and experimenal study that should next follow.

Although not all the introduced plants are mentioned, it can be seen that they are far from numerous; in fact, they play comparatively little part as yet in the economy of the association. Nor is it to be expected that they will greatly increase either in number or aggressiveness under the present condition of affairs. The truth seems to be that a balance has been reached and that a plant association unknown in primitive New Zealand is now well established.

– 174 –

(4.) Swamp.

Every transition from lake to tussock grassland is to be seen at Cass; but our data is quite insufficient for even a superficial treatment of the subject, consequently our remarks are confined to swamp But, even in this regard, the details, on which so much depends, of the relation between the species and the depth of the water cannot be given. By “swamp” we mean an association growing in a habitat where water remains permanently, or at least during the greater part of the year, on the surface of loose, frequently peaty, soil, while it is not too deep to inhibit the presence of plants rooting in the ground. This formation varies considerably in character in different parts of New Zealand, while, except forest, no formation has been so greatly altered by man. The Cass swamps, then, since they are virtually in their primitive condition, stand as most important natural objects, to be jealously guarded and assiduously studied. The piece of swamp close to the biological station supplies a fine example of the gradual change from glacial lake to dry ground by way of swamp.

The swamp belongs to the class “reed-swamp,” and to that association where Typha angustifolia var. Muelleri is dominant—raupo (Typha) association.

The vegetation exhibits a fairly well-marked series of girdles which are in harmony with the depth of the water, but the effect is to some extent masked by streams of running water passing through the swamp, and by non-uniformity of depth in places between the centre and margin. Speaking generally, there is a piece of open water in the centre of the swamp too deep as yet for occupation by swamp-plants; next comes a girdle of raupo (Typha angustifolia var. Muelleri); next a girdle of Carices and other plants, then the shore girdle, which is subject to periodical submersion; and, finally, boggy ground, usually beyond the reach of flooding. Our notes do not permit exact details of the composition, or even the limits, of these girdles.

The Carex girdle has as its dominant species the niggerhead, Carex secta, which species in places invades the Typha area. Growing with the above Carex is more or less Phormium tenax. Where the water is shallower the grass-like Carex Gaudichaudiana is easily dominant, forming a broad girdle With it are other plants—e.g., Epilobium pallidiflorum—but we have no precise details.

At the margin of the swamp proper there is a considerable assemblage of species, which, though here taken together, do not all grow under the same conditions The following may be mentioned: Carex Gaudichaudiana (stunted), C. ternaria, Schoenus pauciflorus, Juncus polyanthemos var, Luzula campestris var., Rumex flexuosus, Montia fontana, Potentilla anserina var. anserinoides, Geranium microphyllum, Viola Cunninghamii, Halorrhagis micrantha, Hydrocotyle novae-zelandiae var montana, Oreomyrrhis andicola var, Mazus radicans, Plantago triandra, Asperula perpusilla, Celmisia longifolia var., Gnaphalium paludosum.

Schoenus pauciflorus plays an important part in the marginal physiognomy of the swamp, since in no few places it forms almost pure patches This sedge of the tussock-form is especially characteristic of shallow gullies, where it grows in company with Sphagnum and formas a distinct bog association. Common species of such a habitat are: Blechnum penna marina, Poa caespitosa var., Hierochloe redolens, Carex ternaria, C. Gaudichaudiana var., Viola Cunninghamii, Hydrocotyle novae - zelandiae var. montana, Aciphylla squarrosa var., Asperula perpusilla, Wahlenbergia albomarginata, Celmisia longifolia var., Olearia virgata var.

– 175 –

(5.) Rock Associations.

Dry rocks can only support a scanty assemblage of plants consisting of highly specialized, drought-enduring forms, or of species that can epharmonically alter their form. Such, to cite some of the species belonging to the first category found in such a habitat at Cass, are: Colobanthus acicularis, Pimelea Traversii, Veronica epacridea, V. tetrasticha, and Helichrysum Selago var. But the fissures, hollows, and crevices of rock readily become filled with soil, and so provide a habitat more easily colonized, where many plants belonging to less xerophytic stations can thrive. When such positions are in the shade, or are so situated as to receive an abundant supply of water, there will be a rich florula, but many of the species will have little to do with rock conditions The following lists include only those species which are either true rock-plants or are commonly found on rocks where soil has accumulated, though also belonging to other associations.

(a.) Montane Rock.

Most of the species belong equally to other associations. The following list does not discriminate between the plants of dry rocks and those of moister rooting-places: Blechnum capense var., Poa Colensoi var., P. caespitosa var., Festuca novae-zealandiae, Luzula campestris var., Colobanthus acicularis, Geum parviflorum, Carmichaelia subulata, Coriaria sarmentosa var., Discaria toumatou, Aristotelia fruticosa, Viola Cunninghamii, Hymenanthera dentata var. alpina, Pimelea Traversii, Leptospermum scoparium var., Epilobium melanocaulon, Angelica montana, Anisotome aromatica var., Corokia Cotoneaster, Gaultheria rupestris var., Styphelia Fraseri, Dracophyllum uniflorum var., Veronica salicifolia var. communis, V. Traversii var., Coprosma propinqua var, Wahlenbergia albomarginata, Olearia avicenniaefolia var. Helichrysum bellidioides var., H. Selago var., Cassinia fulvida var. montana.

(b) Subalpine Rock.

Here rocks on Mount Sugarloaf are alone considered. The habitat is dry, and frequently wind-swept The altitude is not sufficient for certain species more or less common on rocks of the neighbouring higher mountains. The following is a list of the species: Podocarpus nivalis, Aciphylla Colensoi, Anisotome aromatica var. stunted, Dracophyllum romarinifolium, Styphelia Colensoi, Suttonia nummularia, Veronica epacridea, V. tetrasticha, Celmisia spectabilis, Senecio Bidwillii. Almost certainly Colobanthus acicularis and Pimelea Traversii are present, but they are not mentioned in our notes.

(6.) River-bed.

Broad stony river-beds, over which anastomosing streams wander, are a frequent feature of the South Island of New Zealand. They carry a fairly uniform vegetation, which exhibits certain well-defined phases in harmony with topographical and biological changes. The bed proper is that portion of the stony area occupied by the streams, and liable at any time to be flooded and its plant-covering to be eradicated; it is, in fact, a habitat of extreme instability. Popularly included in the term “riverbed” is the more stable ground, formerly the flood-plain, on either side of the bed proper, and this we likewise include in our treatment of riverbed. Throughout the South Island the vegetation under consideration has a similar life-history.

First of all, there is the extremely open plant-covering of the unstable bed, which, owing to the instability of the substratum, does not pave the

– 176 –

way for future development, but is a distinct beginning and end in itself—i.e., it is a plant formation closely related to but distinct from the later-formed river-bed associations. All the same, it is always present, since when destroyed at one spot it is being renewed elsewhere.

This primary community may be designated the Epilobium association, since one or other species of this genus is the first-comer, thanks to the wind-borne seeds, their rapid germinating-power, and the quick development of the seedlings Also, there is present more or less of the mat-forming Raoulia tenuicaulis. The characteristic species of Epilobium for montane river-bed are the erect-branching E. melanocaulon and one or more of the varieties of the mat-forming E. pedunculare, a polymorphic species.

As the bed becomes stable—i.e., as low terraces are built by the river with their surfaces beyond the reach of the highest floods—another association makes its appearance, and there is now a gradual procession of events, according to climate and altitude, from an association where mat and low cushion plants dominate to low tussock grassland, shrubland, or even forest.

Coming now to the vegetation of the river-bed in close proximity to the biological station—namely, that of the River Cass—we are face to face with some apparent discrepancies in the above general remarks. This river, which rises in the Craigieburn Mountains, is only some eight miles in length. Its upper portion, flowing through a forest-clad gorge, is of a more or less torrential character, and does not come within the scope of this paper. The lower part of the river flows through its ancient floodplain, which at the widest part is some two miles across. Here the true river-bed is remarkably broad in proportion to the water which it carries, so that it is far more stable than the habitat in general, and in consequence the primary and secondary plant-communities are not sharply defined, and there are far more species on that part of the bed, still liable to flood, than is usually the case in such a habitat. All the same, the actual procession of events is as detailed above, though it is not so evident as on many river-beds.

The vegetation of the unstable river-bed, now about to be described, is not one association, but a combination of the true primary Epilobium association and the succeeding Raoulia association. The most important members are Epilobium melanocaulon, Raoulia tenuicaulis, and R. australis var. On portions of the substratum not swept bare by water for some time there may be tussocks of Festuca novae-zealandiae, and the introduced Holcus lanatus, Rumex Acetosella, and Cerastium triviale.

On river-terrace recently formed, but which is of sufficient age to have enabled plant-colonists to become established, in addition to the species already mentioned, the following are present: Muehlenbeckia axillaris forming circular wiry interwoven mats, the introduced Sagina procumbens, several varieties of Acaena inermis and A microphylla forming rounded mats, the introduced Trifolium repens, rosettes of Geranium sessiliflorum var. glabrum, Discaria toumatou (here prostrate), Raoulia Haastii forming dense green cushions, and the introduced Hypochoeris radicata On such a young terrace the vegetation is liable to rapid destruction when attacked by a stream which has changed its course.

On long-established terrace occupation by plants for a considerable period has added humus to the soil, so that a more favourable station for vegetation has gradually developed, thanks to the plants themselves,

Picture icon

Cushion of Myosotis uniflora Hook f growing on stony river-bed at Cass, showing the relatively large flower

– 177 –

especially the species of Raoulia and to the moss Racromitrium lanuginosum. These Raoulias, according to the species, either collect more or less sand and silt between their branches or the dead parts remain as peat within the cushion. In any case, a seed-bed suitable for the well-being of various species of other plants is provided, which, as they develop, in no few instances kill their host, whose remains, however, add humus to the soil. Raoulia tenuicaulis early on is thus suppressed. The following, in addition to most of the species already noted, are important members of this stage of riverbed: Carmichaelia Monroi, forming open semi-cushions of short, rigid, flat, leafless stems; the introduced Trifolium arvense; Raoulia lutescens, forming dense, low, flat, silvery cushions in great profusion; and the grey, half-dead-looking Helichrysum depressum, a low shrub with minute appressed scale-like leaves. The rare herb Myosotis uniflora occurs here and there; it forms small moss-like cushions (see Plate XVI), and produces rather large pale-yellow flowers. As the humus-content of the soil increases, tussocks of Festuca novae-zealandiae enter into the association, and by degrees low tussock grassland is established.*

(C.) List of Species Belonging to The Associations Dealt with in This Paper.

(Unless the contrary be stated, the species are evergreen. Introduced species marked with an asterisk.)
Name of Species, &c Growth-form Association.
Filices
Cystopteris novae-zealandiae J. B. Armstrong (= C. fragilis Cheesem. non Bernh) Small tufted fern River-terrace and debris scrub.
Hypolepis millefolium Hook. Summer-green creeping fern Forest; river - terrace and debris scrub
Blechnum (Lomaria) capense Schlcht. var. Creeping fern Forest; montane rock.
—— (L) penna marina (Poir) Kuhn (=Lomaria alpina Spreng.) Small creeping fern Forest; river - terrace and debris scrub; tussock grassland; bog; Cassinia open shrubland.
Asplenium Richardi Hook. f. Small tufted fern Coarse debris; montane rock.
—— flabellifolium Cav. Small prostrate fern River-terrace and debris scrub.
Polystichum vestitum Forst. f (= Aspidium aculeatum Swartz var vestitum Hook. f.) Robust tufted fern Forest; river - terrace and debris scrub montane rock.
Ophioglossum coriaceum A. Cunn. (= O lusitanicum L. in part as defined in Man N.Z Flora) Summer-green fern Tussock grassland.
Botrychium ternatum Sw var. " "

[Footnote] *This is a very brief account of a most interesting series of associations, but a much more detailed description is in course of publication by Foweraker. In the meantime those who desire further particulars may consult L. Cockayne's paper, “On the Peopling by Plants of the Subalpine River-bed of the Rakaia, Southern Alps of New Zealand” (Trans. Bot. Soc. Edin., vol 24 (1911), p. 104), where the ecological conditions of river-bed are discussed, a synopsis of growth-forms furnished, and the ecology of the Raoulia-form dealt with.

– 178 –
Lycopodiaceae.
Lycopodium Selago L. Erect branching herb Forest.
—— scariosum Forst. f. Prostrate woody plant. River-terrace and debris scrub.
—— fastigiatum R. Br. Creeping erect woody plant Forest; tussock grassland; Cassinia open shrubland
Taxaceae
Podocarpus nivalis Hook. in two vars.
  • (i)

    Mat-forming shrub

  • (ii)

    Erect bushy shrub

Open debris shrubland akin to fell-field.
Typhaceae
Typha angustifolia L. var Muelleri (Rohrb.) Graebn. Tall reed - like creeping summer-green herb Swamp.
Potamogetonaceae.
Potamogeton Cheesemanii A. Benn. Water-herb with floating leaves Swamp in open water.
Gramineae
Hierochloe redolens (Forst. f) R. Br. Tufted grass Bog; montane rock; river - terrace and debris scrub.
*Anthoxanthum odoratum L " Swamp; tussock grassland.
Dichelachne crinita (Forst. f. Hook. f. Tall tufted grass . Tussock grassland; river-bed.
Danthonia semiannularis R. Br var. setifolia Hook. f. Small tufted grass . Tussock grassland.
*Aira caryophyllea L Small annual grass Tussock grassland; river-bed.
*Holcus lanatus L Tufted grass . Tussock grassland; margin of swamp.
Triodia exigua T. Kirk Turf - forming creeping grass Triodia subassociation of tussock grassland.
*Dactylis glomerata L. Tussock grass Tussock grassland
Poa caespitosa Forst. f. var " Tussock grassland; river-bed; Discaa thicket.
—— Colensoi Hook f. var . Small tussock grass . Tussock grassland; Cassinia open shrubland, Discaiia thicket.
—— sclerophylla Berggr . Small, rigid, tufted grass On fine debris.
*—— pratensis L. Turf - forming creeping grass Tussock grassland.
Festuca novae-zealandiae (Hack) Cockayne (= F. ovina L. var novae-zealandiae Hack.) Tussock grass . Tussock grassland; river - bed; Cassinia open shrubland.
Agropyron scabrum (R. Br.) Beauv. var. Tufted grass. Tussock grassland.
– 179 –
Name of Species, &c. Growth-form. Association.
Cyperaceae.
Eleocharis acuta R. Br. Creeping rush-like plant Swamp.
Schoenus pauciflorus Hook. f. Slender rush-like tussock Swamp; bog.
Carex diandra Schrank (=C. teretiuscula Good.) Creeping grass-like sedge Swamp.
—— secta Boott Tussock grass-like sedge with stout “trunk” "
—— Gaudichaudiana Kunth var. Creeping grass-like sedge Swamp; bog.
—— ternaria Forst. f. var. " "
—— dissita Sol. var. " Forest.
—— breviculmis R. Br. Small tufted grass - like sedge Tussock grassland.
Juncaceae.
Juncus polyanthemos Buchen. var (= J. effusus L in part as defined in Man. N. Z. Flora) Rush-like tussock Swamp.
Luzula campestris A. DC. var. Small tufted grass-like plant Tussock grassland.
—— sp. As for L. campestris var., but not xerophytic "
Liliaceae.
Phormium tenax Forst. var. Tall iris-like herb of tussock-form Swamp.
Iphigenia novae-zelandiae (Hook. f.) Baker Small summer - green “bulbous” herb Tussock grassland.
Orchidaceae.
Microtis unifolia (Forst. f.) Reichenb. (= M. porrifolia R. Br.) Small summer - green tuberous herb Tussock grassland; river-bed.
Pterostylis mutica R. Br. Small tuberous herb Tussock grassland.
Corysanthes macrantha Hook. f. " Forest.
Fagaceae.
Nothofagus cliffortioides (Hook. f.) Oerst. (= Fagus. cliffortioides Hook f.) Small canopy-tree Forest.
Urticaceae.
Urtica incisa Poir. Low erect herb Forest.
Loranthaceae.
Elytranthe flavida (Hook. f.) Engler Shrubby hemi-parasite. Forest.
—— tetrapetala (Forst. f.) Engler " "
Polygonaceae.
Rumex flexuosus Sol. Herb with tufted leaves Margin of swamp.
Muehlenbeckia complexa (A. Cunn.) Meissn var. Winding liane River-terrace and debris scrub
—— axillaris Walp Mat shrub River - bed; tussock grassland.
*Rumex Acetosella L Creeping herb Tussock grassland; river-bed
Portulacaceae.
Montia fontana L. Branching tufted herb Swamp.
Claytonia australasica Hook. f. Creeping herb Margin of swamp.
– 180 –
Caryophyllaceae.
Scleranthus biflorus (Forst. Hook. f. (two vars.) Cushion herb River - bed; tussock grassland; rock.
Colobanthus Billardieri Fenzl var Small tufted herb Tussock grassland.
—— acicularis Hook. f. Cushion herb. Rock.
*Cerastium triviale Link Annual herb Tussock grassland; river-bed.
Stellaria gracilenta Hook. f. Erect annual? slender herb Tussock grassland.
*Sagina procumbens L. Small mat herb River-bed.
Ranunculaceae.
Clematis australis T. Kirk Tendril-climbing liane River-terrace and debris scrub; open debris shrubland.
—— marata J. B. Armstrong " River-terrace and debris scrub; Discaria thicket; forest.
Ranunculus multiscapus Hook. f. (= R lappaceus Sm. var. multiseapus Hook. f.) Small rosette herb . Tussock grassland.
—— depressus T. Kirk var. Small rosette herb forming patches Swamp.
Cruciferae.
Cardamine heterophylla (Forst. f.) O. E. Schulz var. (= C. hirsuta L. in part as defined in Man. N.Z. Flora) Slender erect branching herb Forest, rock.
Pittosporaceae
Pittosporum divaricatum Cockayne (= P. rigidum Hook f. in part as defined in Man. N.Z. Flora) Divaricating shrub River-terrace and debris scrub.
Rosaceae
Rubus australis Forst. f. var. glaber Hook. f. Scrambling liane Forest, outskirts.
—— schmidelioides A. Cunn var. coloratus T. Kirk " River-terrace and debris scrub.
—— subpauperatus Cockayne " River-terrace and debris scrub; forest.
Geum parviflorum Sm. Rosette herb Shaded montane rock
Potentilla anserina L. var. anserinoides (Raoul) Cheesem. Creeping tufted herb Margin of swamp
Acaena Sanguisorbae. Vahl var. pusilla Bitter Mat herb . River-terrace and debris scrub; forest
—— —— var. pilosa T. Kirk " Tussock grassland; river-terrace and debris scrub; Cassinia open shrubland.
—— hirsutula Bitter? (= A. adscendens Vahl in part as defined in Man. N.Z. Flora) " River-terrace and debris scrub
—— microphylla Hook. f. (one or more vars) " River-bed
—— inermis Hook. f. (several vars) " River-bed, tussock grassland.
– 181 –
Leguminosae.
Carmichaelia subulata T. Kirk Erect leafless rigid shrub Tussock grassland; rock; river-bed; river-terrace and debris scrub.
—— uniflora T. Kirk. Mat shrub Stony dry part of tussock grassland.
—— Monroi Hook. f. Open rigid cushion shrub River - bed; tussock grassland where especially dry.
*Ulex europaeus L. Spinous dense shrub River-bed.
*Trifolium repens L Creeping turf-making herb Tussock grassland; river-bed.
*—— arvense L Erect annual herb River-bed.
*—— dubium Sibth Small prostrate herb Tussock grassland.
Geraniaceae.
Geranium microphyllum Hook. f. Prostrate herb Bog; river-terrace and debris scrub.
—— sessiliflorum Cav. var. glabrum Kunth (possibly there are two vars.) Rosette herb Tussock grassland; river-bed; rock.
Oxalidaceae.
Oxalis corniculata L. var. Mat herb Tussock grassland.
Coriariaceae.
Coriaria sarmentosa Forst. f. var. (= C. ruscifolia of Man. N.Z. Flora in part) Creeping summer-green semi-woody plant Rock; river - terrace and debris scrub.
Rhamnaceae.
Discaria toumatou Raoul
  • (i)

    Semi - divaricating spiny shrub

  • (ii.)

    Prostrate shrub

  • (i.)

    Discaria thicket; tussock grassland; rock; Cassina open shrubland; river-terrace and debris scrub.

  • (ii.)

    River-bed.

Elaeocarpaceae.
Aristotelia fruiticosa Hook. f.
  • (i.)

    Divaricating shrub

  • (ii.)

    Bushy shrub

  • (i.)

    River - terrace and debris scrub, and open debris shrubland.

  • (ii.)

    Forest; river - terrace and debris scrub.

Violaceae.
Viola Cunninghamn Hook f. Rosette herb Bog; tussock grass-land; Cassina open shrubland.
Hymenanthera dentata R. Br. var. alpina T. Kirk. (This may be merely an epharmonio form of H. crassifolia Hook. f.) Open cushion, almost spinous shrub Rock; river - terrace and debris scrub; scrub; forest, but here of open form; Cassinia open shrub-land; open debris. shrubland.
– 182 –
Thymelaeaceae
Pimelea Traversi [ unclear: ] Hook. f. Small dense shrub Rock; open debris shrubland
—— prostrata (Forst f) Willd. var. repens Cheesem Mat shrub River - bed; tussock grassland
—— —— var erect Small erect dense shrub Tussock grassland; Cassinia open shrub-land
Drapetes Dieffenbachii Hook Turf-making semi-woody plant Shady tussock grassland
Myrtaceae
Leptospermum scoparium var Forst. Somewhat fastigiate ericoid shrub Manuka thicket; Cassinia open shrubland; open debris shrubland
Onagraceae.
Epilobium pallidiflorum Sol Tall erect herb Swamp
—— pubens A. Rich . Erect herb, woody at base River-terrace and debris scrub; rock
—— pictum Petrie Erect herb River-terrace and debris scrub.
—— chloraefolium Hausskn " River-terrace and debris scrub
—— rotundifolium Forst f Erect or semi-prostrate herb Forest.
—— pedunculare A Cunn (several vars) (= E [ unclear: ] nummularifolium R. Cunn var pedunculare Hook. f.) Mat herb River - bed; tussock grassland
—— melanocaulon Hook . Erect branching herb, woody at base River-bed; rock.
—— microphyllum A. Rich Erect branching herb with woody base River-bed
—— elegans Petrie? (This may be an undescribed species or a var of E. novae-zealandiae Hausskn) Low erect herb Tussock grassland.
Halorrhagaceae
Halorrhagis micrantha R. Br . Low prostrate herb Margin of swamp.
Myriophyllum elatinoides Gaud? Partly submerged water-plant Swamp
Umbelliferae
Hydrocotyle novae-zelandiae DC var montana T. Kirk Creeping herb Tussock grassland; margin of swamp
Schizeilema Hookeri (Drude) Domin (= Azorella trifoliata Benth & Hook) Slender creeping herb Forest
—— (Az) [ unclear: ] nitens (Petrie) Do [ unclear: ] Oreomyrrhis andicola Endl var Colensoi (Hook. f.) T. Kirk. (There are probably other vars, or, it may be, all are merely dependent on environment) Slender matted herb Low herb Margin of swamp Tussock grassland.
Aciphylla Colensoi [ unclear: ] Hook. f. Yucca-like spinous rigid herb Rock
– 183 –
Ac [ unclear: ] phylla squarrosa Forst. var. Yucca-like spinous rigid herb Tussock grassland; rock.
Anisotome filifolia (Hook. f.) Cockayne and Laing (= Ligusticum [ unclear: ] filifolium Hook f.) Grass-like rosette herb River-terrace and debris scrub (probably also grows on stony debris); open forest.
—— aromatica Hook. f. var (= Ligusticum aromaticum Hook. f.) (This aggregate species badly needs splitting up into its several very distinct varieties) Low rosette herb Tussock grassland; rock.
Angelica montana (Forst.) Cook-ayne (= A. Gingidium Hook. f.) Stout herb with leaves in open rosettes Rock; side of stream in scrub.
Cornaceae.
Corokia Cotoneaster Raoul Divaricating shrub River-terrace and debris scrub; rock.
Ericaceae
Gaultheria depressa Hook. f. (= G. antipoda Forst f. var depressa Hook. f.) Low creeping matted shrub Shady tussock grassland; Cassinia open shrubland.
—— rupestris R. Br. var. Small erect or depressed stiff shrub Rock; river - terrace and debris scrub; open debris shrubland.
Epacridaceae.
Styphelia (Cyathodes) acerosa Sol var. Prostrate ericoid shrub Rock.
—— (Cy) Colensoi (Hook. f.) Diels Mat shrub Rock; tussock grassland; Cassinia open shrubland.
—— (Leucopogon) Fiaseri (A. Cunn.) F. Muell Low erect shrub with creeping stem Tussock grassland; rock; river-bed; Cassinia open shrubland.
Dracophyllum longifolium (Forst) R. Br. var. Erect fastigiate shrub Forest-margin; open debris shrubland.
—— uniflorum Hook. f. (probably D. acicularifolium (Cheesem) Erect branching shrub River-terrace and debris scrub; Cassinia open shrubland.
—— rosmarinifolium (Forst. f.) R. Br Prostrate rigid shrub Rock.
Pentachondra pumila (Forst. f.) R. Br. Small cushion shrub Tussock grassland, in shade.
Myrsinaceae
Suttonia nummularia Hook. f. (= Myrsine nummularia Hook. f.) Prostrate trailing shrub Rock
– 184 –
Gentianaceae.
Gentiana tenuifolia Petrie Erect slender herb River-terrace and debris scrub.
—— sp. (This seems distinct from any species of Gentiana with which we are acquainted, but no flowering material is available) Rosette herb Tussock grassland.
—— corymbifera T. Kirk var . " Tussock grassland.
Boraginaceae
Myosotis uniflora Hook. f Small cushion herb, River-bed.
—— australis R. Br. var. (probably quite distinct from the Australian plant) Erect loose-rosette herb Tussock grassland; open debris shrubland
Scrophulariaceae Mazus radicans (Hook. f.) Cheesem. Creeping mat herb Margin of swamp.
*Verbascum Thapsus L. Large - leaved rosette plant Where scrub has been burned.
Veronica salicifo [ unclear: ] a Forst f. var communis Cockayne Bushy shrub . River-terrace and debris scrub
—— Traversii Hook. f. var. Ball-like shrub River-terrace and debris scrub; Discaria thicket; rock.
—— buxifolia Benth var. odora T. Kirk " Scrub of wettish ground
—— pimeleoides Hook. f. var minor Hook. f. Very low creeping shrub Tussock grassland
—— tetrasticha Hook. f Very low shrub, with reduced imbricating leaves Rock.
—— epacridea Hook. f. var Prostrate rigid shrub "
—— Lyallii Hook. f. var. Slender mat semi-woody plant Rock; margin of stream in scrub.
—— Bidwillii Hook. f. Ditto Rock.
Plantaginaceae
Plantago spathulata Hook. f. Stout flat rosette herb. Tussock grassland
—— triandra Berggr. Small flat rosette herb. Margin of swamp
Rubiaceae.
Coprosma parviflora Hook. f. var Divaricating shrub River-terrace and debris scrub, forest; rock
—— brunnea (T. Kirk) Cockayne (= C. acerosa A. Cunn. var. brunnea T. Kirk) Rigid prostrate creeping shrub Rock (should also be on river-bed)
—— rugosa Cheesem Divaricating shrub . River-terrace and debris scrub
—— propinqua A. Cunn. var " River-terrace and debris scrub, forest
—— [ unclear: ] nar [ unclear: ] folia Hook. f. Bushy shrub Forest.
—— cuneata Hook. f. var Divaricating shrub . Rock
—— microcarpa Hook. f.? . " Forest.
—— Petriei Cheesem (two vars with differently coloured drupes) Mat or cushion shrub . River - bed; tussock grassland.
– 185 –
Galium umbrosum Sol. Straggling slender herb River-terrace and debris scrub; forest.
Asperula perpusilla Hook. f. Very low slender herb Tussock grassland; margin of swamp.
Nertera Balfouriana Cockayne Small creeping matted herb Margin of swamp.
Campanulaceae.
Wahlenbergia albomarginata Hook. (= W. saxicola Cheesem. non A. DC.) Small creeping rosette herb Tussock grassland; bog; rock.
Compositae.
Brachycome Sinclairii Hook. f. Low creeping rosette herb Tussock grassland.
Olearia avicenniaefolia Hook. f. var. Tall shrub of tree-composite form River-terrace and debris scrub; rock.
—— virgata Hook. f. var. Divaricating shrub River-terrace and debris scrub; bog.
Celmisia spectabilis Hook. f. Mat or low cushion rosette herb, woody at base Tussock grassland; rock; open debris shrubland.
—— Lyallii Hook. f. Stout erect rosette herb Open debris shrubland.
—— longifolia Cass var. Rather grass - like erect herb Tussock grassland; margin of swamp; Cassinia open shrubland
Vittadinia australis A. Rich. Small erect branching herb, woody at base Tussock grassland.
Gnaphalium Traversii Hook. f. Low, creeping, woolly mat herb "
—— paludosum Petrie Slender tufted herb Margin of swamp.
Raoulia australis Hook. f. (two vars.) Dense silvery semi-woody mat plant River-bed.
—— lutescens (T. Kirk) Beauv. (= R. australis Hook. f. var lutescens T. Kirk) Dense silvery mat or low cushion "
—— tenuicaulis Hook. f. Dense mat or semi-woody plant "
—— subsericea Hook Dense mat-or turf-making semi-woody plant Tussock grassland; Cassinia open shrubland.
—— Haastii Hook. f. Dense green cushion or semi-woody plant River-bed.
Helichrysum bellidioides (Forst. f.) Willd. var Creeping slender mat herb, woody at base Tussock grassland; bog; rock.
—— depressum (Hook. f.) Benth. & Hook. f. Low, rigid, cupressoid shrub River-bed.
—— Selago (Hook. f.) Benth. & Hook. var. Low, stout, cupressoid shrub Rock.
Cassinia fulvida Hook. f. var. montana Cockayne [ unclear: ] var. nov. ined. Erect bushy ericoid shrub Cassinia open shrubland; tussock grassland.
Cotula squalida Hook. f. Creeping turf - making herb River-bed; dry tussock grassland.
Erechtites glabrescens T. Kirk Tall erect herb Forest; river - terrace and debris scrub.
*Hypochoeris radicata L. Rosette herb Tussock grassland; river-bed; rock.
– 186 –
Senecio bellidioides Hook. f. var. glabratus T. Kirk. (Many individuals are tomentose on the under surface of the leaf, but we hardly think another variety should be established for this reason) Low rosette herb . Tussock grassland
—— Lyallii Hook. f. Grass-like rosette herb Scrub on bank of streams.
—— lautus Forst. f. var. montanus Cheesem. Small rosette herb Open debris shrubland.
—— Bidwillii Hook. f. Stout shrub of tree-composite form Rock.
Microseris scapigera Sch Bip. (= M. Forsteri Hook. f.) Low rosette herb Tussock grassland; Cassinia open shrubland.
Taraxacum magellanicum Comm. (= T. officinale Cheesem non Wigg and T. glabratum (Forst. f.) (Cockayne) " Tussock grassland; margin of swamp

[Footnote] † This form is very similar to that of the well-known Rhododendron ponticum.