Art. XXXIV.—The Vegetation of Banks Peninsula, with a List of Species (Flowering-plants and Ferns). By Robert M. Laing, B.Sc. [Read before the New Zealand Institute, at Christchurch, 4th–8th February, 1919; received by Editor, 2nd April, 1919, issued separately. 19th August. 1919.] The Indigenous Vegetation of Banks Peninsula. Physiographic. Banks Peninsula, on the east coast of the South Island of New Zealand, is such a well-defined and isolated area that it is remarkable that it has not received more attention from botanists. The list of papers at the end of this article shows how scanty has been the botanical work done on this group of hills. Indeed, there is at present no reliable li t of the species occurring there. This is the more to be regretted as no portion of the area except the coastal cliffs and the salt marsh now remains in its original condition; and no doubt even on the cliffs some introduced species of plants are to be found. The complete destruction of the forest and the annual burning of the tussock areas have so altered the plant associations that it is difficult to reconstruct them, even in imagination, with accuracy. Some species are now undoubtedly lost, many more have been introduced from without, and the relative numbers of those present have been, of course, totally reproportioned. Any attempt to describe in detail the distribution of species over the area before the arrival of the white man must of necessity fail, but the rough outlines of that distribution can still be determined. The following paper therefore attempts to give that information, and to provide a list of the indigenous species that should be useful to future students of the area. It is not necessary to describe the chief physiographic features of the area, as this has already been done in the papers of von Haast, Hutton, and Speight, which are readily accessible Suffice it to say that Banks Peninsula stretches out to the south-east from the centre of the eastern side of the Canterbury Plains. It is oval in shape, and about thirty-five miles long and twenty wide. It consists of a congeries of hills rising at the centre in Mount Herbert to a height of just over 3,000 ft., and in Mounts Sinclair and Fitzgerald to a slightly less height. From these and other peaks long ridges with steep sides run out in all directions, enclosing occasionally narrow flats containing several hundred acres of land. Beyond the flats and between the outer ridges are the smaller bays. On the seaward side the ridges terminate in cliffs 300 ft. to 500 ft. high; and on the landward side slope down to the plains, cliffs being absent. Two large harbours, on the sites of old volcanic calderas, break into the hills, and are surrounded by steep walls which rise in rocky cliffs and escarpments to the height of 2,000 ft. in Akaroa Harbour, and somewhat less in Lyttelton Harbour. The longest valleys are the Little River and Kaituna Valleys, each some seven miles in length, both at one time, throughout part at least of their lengths, arms of the sea. To the south-west lies a large, shallow, brackish mere—Ellesmere—formed by the blocking-up of the mouth of the Selwyn River. This action is due to the shingle drifted up the coast from the mouth of the Rakaia River by the southern current.

Some Localities Not Marked on Map. Allandale, between Governor's Bay and Teddington. Aylmer's Valley, one of the valleys behind Akaroa Balgueri Valley, behind Akaroa Township. Barry's Bay, near the head of Akaroa Harbour. Castle Rock, also known as Mount Herbert Peak. Charteris Bay, at the foot of Castle Rock. Caton's Bay, on Lake Forsyth, near Little River. Damon's Bay, north of Akaroa East Head. Dover Castle, rocky wall overlooking Heathcote Valley. Flea Bay, north of Damon's Bay. Gollan's Bay, at the foot of the Zigzag, Lyttelton-Sumner Road. Island Bay, between Akaroa Heads and Peraki. Long Bay, There are three Long Bays on the peninsula, but indications are given in the text as to the one referred to. Long Lookout Point, east of Little Akaloa Bay. Ohinitahi, on Lyttelton Harbour, between Governor's Bay and Teddington. Okute Valley, running from Saddle Peak to Little River. One Tree Hill, between Lyttelton and Port Levy. Otahuna Valley, running from Cooper's Knobs to the plains. Rapaki, Maori settlement between Lyttelton and Governor's Bay. Stony Bay: The one referred to is between Little Akaloa and Okain's. There are two bays of the name on the peninsula. Taylor's Mistake, between Sumner and Lyttelton Heads. Teddington, at the head of Lyttelton Harbour. Tikao Bay, opposite Akaroa, on the harbour. Timutimu, west head of Akaroa Harbour Waikerikikeri, bay south of Le Bon's, and popularly known as Hickory. Wainui, between Tikao Bay and the Heads.

We are dealing, then, with an area which consists of long, moderately steep hill-slopes, radiating out from the central heights, It is broken on its outward sides by harbours, bays, and valleys, with steep cliffy sides. There is comparatively little flat land. Alpine shingle-slips are completely wanting, and screes are of rare occurrence. Bogs and swamps are of very small extent, and consist of the damp spots in the neighbourhood of springs and streams. Salt marshes and meadows, however, abound, chiefly in the neighbourhood of Lake Ellesmere. The highest hills rise above the forests into subalpine grasslands. On three sides—the north, south, and east—the peninsula is surrounded by the ocean; but on the western side it meets the Canterbury Plains The attached map will give the names of the chief places referred to, and a list is provided with the situations of the less known minor localities. Climatic. The climate of the peninsula may be described as typically insular and warm-temperate. It differs somewhat from that of the Canterbury Plains, which is more continental in type. Thus the plains are much colder on winter nights than the adjacent hills. The temperature of Cashmere Hills on frosty nights is from 3° C. to 5° C. higher than that of Christchurch, and on the hills at Redcliffs and Sumner it is still warmer. There ice is rarely seen. The lowest temperature that may be observed on the Cashmere Hills is about —8° C., but winters pass in which it does not fall below —4° C. or —5° C. Even in midwinter the days are often bright and warm, and the temperature in the lower valleys is as high as 15° C. in the shade. Frosts may occur on the plains in any month of the year, but on the neighbouring hills they are almost unknown between September and April. Thus in the beginning of November, 1918, an unusually severe frost (—7° C.) occurred in Christchurch. This was so slightly felt on the hills that tomatoes and potatoes remained unchecked and almost untouched by it. The conditions on the hilltops have been but little studied, though, of course, the temperatures there must be considerably lower both in summer and winter than those at the foot. The temperature range through the year on the lower portion of the hills is comparatively small, but reliable records are difficult to obtain. Probably the difference of average temperature is not more than 8° C. or 9° C. between a month of winter and one of summer, though the maximum temperature in summer is often comparatively high, and the thermometer may even rise above 33° C. In the warm, sheltered valleys on the north side of the peninsula the cond tions must approach the warm-temperate, and this is shown in the fact that many typical North Island plants find here their southernmost limit. (A list is appended, pp. 369–70.) The rainfall varies in different localities, but as comparatively few records have been taken on the peninsula it is impossible to give detailed results. However, the average probably varies from about 25 in. (Convalescent Home, Cashmere Hills) to about 50 in. or 60 in. (on the top of Mount Herbert). A series of careful observations extending over nineteen years (1899–1918) at Pigeon Bay, taken by Mr. E. Hay, give an average of 29.5 in., with a maximum of 39.5 in in 1913 and a minimum of 16 in. in 1915. The rainfall in the outer portions of the area between Pigeon Bay and Akaroa is no doubt higher than this. In the latter place the records show an average of 45 in. Between Timutimu Head and Birdling's Flat there is again a reduction in the rainfall, for the Akaroa Hills cut off

much of the easterly rain. This district therefore reproduces largely the flora of the Lyttelton Hills, but, having a southerly aspect, no doubt is neither quite so hot nor so dry as they are. Snow falls every year on the hilltops, and on Mount Herbert usually lies for some weeks, but at the base of the hills it does not often fall, and usually does not lie for more than a few hours. The winter of 1918 was exceptional, when snow fell to a depth of 10 in. oven at the foot of the hills, and the conditions above 1,000 ft. were truly alpine. Such falls do much damage in the forest in breaking down trees and branches, though but little permanent harm results if frosts do not follow. Next to the rainfall and temperature the most important climatic factor is the wind. The prevailing wind is from the north-east. This in summertime is usually a sea-breeze, and then brings no rain, but when part of a cyclonic system it frequently brings continuous though often light rains lasting over many hours. The north-east wind passes into the north-west wind, which is much less frequent, and is hot and dry. Only on the rarest occasions does it bring a scanty shower of rain. It has, as will be seen later, a most important effect on the distribution of the plants within the area. The south-wester is a cold, wet wind, bringing much rain, and determines largely the vegetation on the cliff-faces exposed to the south. The south-easter is, on the northern side of the peninsula, a somewhat rare wind, of a gusty character—often stormy and occasionally bringing heavy rains. It is often deflected as an easterly or north-east wind. The higher rainfall of the outer portion of the peninsula is largely due to it. The rainfall is very irregular in its distribution over the year. The months of December to March are usually dry, but exceptions occur. During this period droughts are not uncommon, particularly on the north-west faces of the peninsula, and often affect the vegetation. This has tended to produce a distinctly xerophytic type of vegetation on the Lyttelton Hills, though elsewhere it tends to the mesophytic. Changes in the Plant Covering. It has already been stated, and cannot be too strongly insisted on, that here—as in so many other places in New Zealand—we are dealing with a vegetation that has during the last seventy or eighty years undergone, and is still undergoing, immense changes. The greatest of these is, of course, the disappearance of the forest. Captain Thomas, Agent for the Canterbury Association, reporting on the 15th May, 1849, estimated that half of the area of Banks Peninsula—viz., 134,000 acres—was forest; but this probably was an underestimate, as nearly two-thirds of the peninsula must have been forest-covered. Now there is probably nowhere on the peninsula a stand of 300 acres of timber-trees, although there are larger areas of the smaller bush trees, as in the forest at the back of Mount Herbert. The reserves at Peraki Saddle, including some 167 acres, perhaps contain the best specimen of primitive forest now to be seen on the peninsula. Though there are equally fine trees near the Lyttelton—Kaituna saddle, yet here the undergrowth has to a large extent been destroyed by stock. It is unfortunate that what up till recently was the finest forest on the peninsula—viz., the one at Stony Bay—is now being cut out by its owners. Kennedy's Bush Reserve, near Christchurch, unfortunately contains few trees of any magnitude, and its former wealth of tree-ferns has been completely destroyed.

Agents altering Plant-distribution. The primitive plant covering, then, has largely disappeared, and has been replaced by introduced species, particularly by grasses; the balance in the remaining plant associations has been altered, and the conditions under which many of the species live are quite changed. Probably the grasslands of the Lyttelton Hills have been less altered than the forests of the peninsula, for the tussock form still dominates; yet here certainly many changes have taken place and others are still in progress. The chief destructive agents at work where cultivation has not been employed are continual tussock-fires, bush-fires, drought, sheep, cattle, on the Lyttelton Hills rabbits, and on Banks Peninsula hares. On the grasslands the tussock-fires make for the extinction of the ferns, for the reduction in the number of the species, for the replacement of certain grasses by others, for the destruction of isolated shrubs, and for the multiplication of a few plants that can to some extent withstand the action of fire—e.g., Coriaria sarmentosa. On the other hand, bush country when burned and allowed to restock itself reproduces the usual fire weeds; but these are seldom or never allowed to remain, being either destroyed by the hand of man or replaced in old burns by Pteridium esculentum or species of Leptospermum. These changes however, will be considered in greater detail under the various plant formations. It may, however, be noted here that in all the moister bays of the peninsula—that is to say, from Port Levy eastward and southward to Akaroa—cocksfoot (Dactylis glomerata) was sown on the burnt areas, and the result has been the development of an artificial plant association. After the first sowing this grass has replaced itself in a remarkable way, and though reaped for forty or fifty years has continued to produce crops of almost undiminished vigour. Largely owing to the scarcity of casual labour the cocksfoot harvest is now becoming a thing of the past, and dairy-farming is to a large extent replacing it. The cattle are grazed on the cocksfoot, which is the chief ingredient of the pasture lands, though rye-grass (Lolium perenne), timothy (Phleum pratense), and crested dogstail (Cynosurus cristatus) are also to be met with in smaller quantity. Sheep are to be found chiefly on the tussock areas, though even on the cleared forest areas they are in many places being stocked in increasing numbers, owing to the smaller amount of labour involved in looking after them. Further discussion of these economic matters would, however, be out of place here. The changes due to the action of animals on the vegetation, though extensive, are not so wide-reaching as those of fire. Sheep crop certain grasses, and thus prevent them seeding. They also attack other species of plants, such as Angelica montana, Carmichaelia subulata, and even the young spinous leaves of Aciphylla squarrosa. On the other hand, their function in the spreading of plants by means of seeds can hardly be overestimated. The large increase in danthonia (Danthonia pilosa) referred to farther on may be cited in this regard. Other species spread by the same means are Urtica urens, Acaena novae-zelandiae, and Marrubium vulgare (white horehound). Cattle, again, are probably entirely destructive. Phormium soon disappears in their presence, as also does Marrubium. In Pigeon Bay, for example, the upper portions of the hills are in many places grey with horehound, whilst the cattle-country immediately below is perfectly clear. Unfortunately, cattle will not run in the rocky, broken country on

the hill-crests, and here Marrubium has become the worst pest of the district. It remains to be seen whether with time it will work itself out. Probably in the absence of sheep it is but little distributed. Cattle also, as is well known, rapidly destroy the undergrowth in bush country. Rabbits, also, on the area dealt with are probably almost entirely destructive in their action. They reach many places which are either inaccessible to sheep or at least not usually grazed by them. On some of the rocky points of the Lyttelton Hills, which would otherwise be plant-sanctuaries, scarcely a plant is left untouched by them. Shrubs and grasses are all grazed, and only the inaccessible chasmophytes escape. Hares apparently are much less destructive than rabbits, because they are in smaller numbers, do not burrow, and do not frequent cliffs and rocks. The tendency of these agents is not only to reduce the number of indigenous species, but also, as a rule, to limit the numbers of the individuals in the remaining species. Certain species, however, are apparently on the increase as the result of changed conditions. Such are Danthonia pilosa, D. semiannularis, Acaena novae-zelandiae, Coriaria sarmentosa, Cotula squalida, Dichondra brevisepala, and Oxalis cormculata. The two last-mentioned invade even cultivated ground, and I have seen O. corniculata growing about the steps of the public buildings of Christchurch. During a succession of dry seasons the grasses mentioned become common in garden lawns on the hills, and replace temporarily or permanently the shallow-rooted lawn-grasses. In dry seasons, too, rabbits attack the bark of trees, and thus assist the drought in its attack upon the vegetation Melicytus ramiflorus, Schefflera digitata, and Nothopanax arboreum are amongst the first to suffer, particularly when standing in the open. The Plant Associations. Further discussion of such changes may be postponed for the present, and taken when we come to consider shortly the chief of the indigenous plant societies. They may be roughly classified as follows:— (1.) Salt marsh (2.) Salt meadow (3.) Coastal rocksCoastal. (4.) Sand-dunes (5.) Coastal scrub (6.) Tussock-grasslands. (7.) Inland cliff and rock. (8.) Forest. (9.) Lowland scrub and heath. (10.) Subalpine scrub. (11.) Subalpine grasslands. (1) (2) The Salt Marsh and the Salt Meadow. These are developed at the head of the tidal flats, and are well seen at Teddington in Lyttelton Harbour, also on the Sumner Estuary and in a few other places on the peninsula. At Lake Ellesmere there are vast extents of brackish marsh and meadow; but these scarcely come within the scope of this paper. Though some of the species occurring in the portion of the lake adjacent to Kaituna have been listed, no attempt has been made to explore the lake and its borders as a whole. It would undoubtedly form an interesting field for the botanist, and one as yet

imperfectly known. The conditions there are somewhat unusual in New Zealand. However, the ordinary salt marsh and meadow of the Banks Peninsula district do not differ much from those found in adjacent parts of New Zealand, and they have already been described for the Sumner Estuary by Mrs. Jennings (Miss B. D. Cross).* Observations on some New Zealand Halophytes, Trans. N.Z. Inst., vol. 43, p. 545, 1911. At Teddington the association is very similar to that at Heathcote, except that I have seen neither Carex litorosa nor Scirpus maritimus there, though Scirpus maritimus occurs at Ohinitahi, at the head of the bay The first plant to be met with in coming in from the seaward is Salicornia australis, in dense cushions, with a small grass (Atropis stricta) growing through the cushions. At a level several inches higher appear Samolus repens and Selliera radicans, followed by the seaside form of Cotula dioica. By the side of the tidal guts are Juncus martimus and Leptocarpus simplex, the latter in comparatively small quantity. The only shrub present is Plagianthus divaricatus, which is plentiful on the tidal flat and close to the coastal rocks where the shallow tidal flat approaches them. (3) (4) Sand-dunes and Coastal Rocks. There are scarcely any sand-dunes on the peninsula, except those at Sumner, and only small areas of sandy beaches; consequently there are few sand-plants to be noted. So far as I know, Spinifex and Pimelea arenaria, are quite wanting; and Scirpus frondosus, Convolvulus Soldanella, Euphorbia glauca, Carex pumila, and other sand-plants by no means common. On the coastal rocks Mesembryanthemum australe is abundant, with Tetragonia expansa and T. trigyna, often trailing downward for many feet. Rhagodia nutans occurs less commonly, and more rarely still Lobelia anceps and Lepidium oleraceum. Vittadinia australis and Tillaea Sieberiana are abundant here and inland; but I have seen Tillaea moschata in only one place. On the wetter rocks Samolus repens and Selliera radicans reappear, often with the fern usually called Asplenium obtusatum, a species which requires further study. The drier coastal rocks and the adjacent clay banks, as has been pointed out to me by Professor Wall, are the home of certain species of rarer ferns, and produce a little plant association which is as uncommon as it is interesting. In addition to Asplenium obtusatum and A. lucidum the following ferns, though found elsewhere, particularly haunt such situations: Cheilanthes Sieberi, Nothoclaena distans, Polystichum Richardi, Polypodium (Cyclophorus) serpens, and Adiantum affine. The last-mentioned has, of course, been much destroyed by picknickers and pleasure-seekers, and I know only one or two spots at the edge of Lyttelton Harbour where it may be now obtained, though it is common in the less frequented bays of the peninsula. Professor Wall writes thus to me: “There is a very peculiar and special feature of the bluffs and steep slopes surrounding Lyttelton Harbour, which deserves mention. A hard, barren rim of baked clay occurs between the tussock land above and the rock below, in which grow Cheilanthes Sieberi and Nothoclaena distans. Potts† T. H. Potts, Out in the Open, p. 77, 1882. describes the formation very exactly, calling the material ‘cob.’ In many places this rim is now invaded and overshadowed by other vegetation, but the primitive state is still to be seen

here and there. A good example is on the north-eastern side of Quail Island. Practically no plant grows in this sort of place except these two ferns.” (5) The Coastal Scrub The coastal scrub is characterized by such species as Olearia Forsteri, Dodonaea viscosa, and Macropiper excelsum, though any of these may be found inland. Angelica geniculata also occurs in. abundance here, and on the peninsula at any rate is not often found in the interior of the forest. Other common species, which, however, have no claim to be considered distinctly coastal, are Muehlenbeckia complexa, Coprosma robusta, and C. Cunninghamii (showing much leaf-variation); whilst both species of Leptospermum frequently come down almost to the water's edge. At one time also there were groves of karaka, which apparently were to be found from Decanter Bay round to Okain's, and possibly farther. On the open flats, as at Birdling's Flat, Muehlenbeckia complexa in rounded clumps, Discaria toumatou, and Carmichaelia subulata frequently become prominent features in the landscape; but, again, these are not distinctively coastal, though occurring near the sea. (6) The Tussock-grasslands. These were originally found only where the hills were directly exposed to the action of the drying north-west wind, or where they were sheltered from the moist easterlies but met the full strength of the cold south-wester. The easterly winds—the prevailing winds of the peninsula—are much less arid than the north-wester, and less violent and cold than the south-wester; so they have interfered but little with forest growth. On the other hand, the area of exposure to the north-wester is well defined by the tussock belts of the north-western faces of the peninsula. The examination of such a point as Adderley Head well illustrates this contention. The tussock on the Lyttelton side is often withered, while the herbage on the Port Levy side is quite green; the rainfall on both sides must be the same. Wakaroa Head, the eastern point of Pigeon Bay, equally well if not better proves the same point. Here, no doubt, the rainfall is somewhat higher than at Lyttelton Heads, so that it is well above the minimum required for forest-production; yet the forest extended along the eastern side of Pigeon Bay only so far as it was sheltered from the direct action of the north-west wind. The projecting end of the point reproduces exactly the vegetation of the northern slopes of Mount Pleasant, even down to the occurrence of the rare Gymnogramme rutaefolia in both situations, and the less rare but still xerophytic Clematis afoliata. Hence we must expect to find tussock-grasslands wherever the slopes of the hills are exposed to the full violence of the north-wester. This plant association is therefore to be found from Godley Head to Birdling's Flat. As far as Dyer's Pass there are tussock pastures, with scrub only in the gullies; to the westward there is tussock on the open hillsides and headlands, while somewhat heavier forest at one time existed in the valleys. Owing to the clearing of the “bush,” tussock is now found in many places where once there was forest. Natural tussock-grasslands are again to be found on the northern slopes of Mount Herbert and One Tree Hill, on the far side of Lyttelton Harbour, though in the deeper and more sheltered valleys of Purau and Charteris Bay there was forest. In the latter place it has been removed only during the last ten years. The characteristic herbs of the tussock pasture are Poa caespitosa, Danthonia pilosa, Scleranthus biflorus, Oxalis corniculata,

Dichondra brevisepala, Aciphylla Colensoi; and on slopes exposed to seabreezes Cotula Haastii and sometimes Convolvulus erubescens. The most characteristic shrubs are Carmichaelia subulata and Discaia toumatou; whilst amongst rocks Muehlenbeckia complexa, Coprosma crassifolia, Sophora prostrata, and a few other species are occasionally found. Particularly towards the hilltops Hymenanthera crassifolia and Corokia Cotoneaster abound. In most places these grasslands are burnt annually in the late winter or early spring—as soon, indeed, as there is a day on which they will burn. This burning has been done right up to the top of Mount Herbert, and has been carried on for the last fifty or sixty years. When done in the months of July, August, and September the tops only of the Poa caespitosa are burnt; but when fires occur later in the year the tussock is often burnt out and killed. As a result the Poa tussocks become less numerous and remain smaller than they were, and in many places a mat of the wiry Danthonia pilosa forms between them. This is a distinctly aggressive species, no place short of actual rock being too hard for it to grow in. It even invades the cocksfoot lands and occupies the drier ridges there. It has thus proceeded at least as far eastward as Pigeon Bay, where it now occupies lands once covered with more valuable introduced grasses. In the north-west areas it may be the most suitable grass available; but it certainly ought, if possible, to be kept out of the richer cocksfoot lands. Poa Colensoi var. intermedia now exists in comparatively small quantity, and tends more and more to be confined to the hilltops, ledges on rocky faces, and similar situations. Even there, however, on the Lyttelton Hills it is often eaten down by rabbits. This pest is not prevalent on the peninsula, and a rabbit-proof fence between Teddington and Gebbie's Valley helps to prevent them reaching there in large numbers. Festuca rubra, Agropyron scabrum, and Danthonia semiannularis also occur, though in smaller quantities. Between Birdling's Flat and Timutimu Head, on the ridges and flat exposed points, the same tussock formation is repeated. Though the valleys were forested, yet the character of the vegetation is much more xerophytic than on the easterly faces of the peninsula, and the conditions of the Lyttelton Hills, with their vegetation, are closely reproduced. (7) Inland Cliff and Rock. Owing to the structure of the hills, rocky cliffs and faces are abundant, and there is a well-developed and highly specialized rock-vegetation, varying considerably with the altitude and aspect. It may be readily studied on the Lyttelton Hills up to a height of 1,500 ft. On the drier rocks with a northerly aspect are found Hymenanthera crassifolia, Corokia Cotoneaster. Sophora prostrata, and usually near the sea Clematis afoliata, as the most characteristic shrubs. Amongst ferns and herbaceous plants the following occur in such situations: Gymnogramme rutaefolia, Polystichum Richardi, Cheilanthes Sieberi, Chenopodium triandrum, Linum monogynum, Epilobium cinereum, Senecio lautus, and more rarely Rhagodia nutans. A traverse of the hills—a distance perhaps of not more than 50 yards—to the moister rocks on the southern side shows a very different and unique vegetation. The most characteristic shrub is now Veronica Lavaudiana. Metrosideros hypericifolia and Cyathodes acerosa also occur; whilst the ferns already mentioned are replaced by Polypodium grammitidis, and on the highest peaks shrivelled specimens of Hymenophyllum multifidum and occasionally

other species of Hymenophyllum are to be found. The most characteristic plant, however, is a species of Senecio, which, as Wall has shown, between Lyttelton Heads and Gebbie's Pass is S. saxifragoides, but elsewhere is Senecio lagopus in a large and well-developed form. Veronica Lavaudiana is found above 800 ft. only, but Senecio saxifragoides comes occasionally down to sea-level, though usually found above 800 ft. In places also Angelica montana, Anisotome Enysii (?), Earina suaveolens, Libertia grandiflora, and Raoulia glabra become members of the rock association; but these species are to be found on all aspects of the hill, and, with the exception of the plant here called Anisotome Enysii, can scarcely be called distinctive species. This Senecio - Veronica Lavaudiana association is highly characteristic of Banks Peninsula, and is met with nowhere else. Above 1,500 ft. the distinction between the vegetation of rock-faces with northern and southern aspects tends to disappear. Dracophyllum acicularifolium var. uniflorum now becomes a highly characteristic shrub of the cliffs and rocky faces, and the subalpine species, to be described later, begin to appear. To see the lower subalpine element at its best one may go to a cliff on the south-west face of a peak—apparently nameless—between Cass Peak and Cooper's Knobs. This hill is under 1,600 ft., but has an unsheltered exposure towards the south-west, and this probably accounts for the great variety of plants to be found on its face. Amongst the smaller forms may be mentioned Hydrocotyle microphylla, Colobanthus Muelleri, Myosotis pygmaea, Polypodium australe var. pumil, Polypodium grammitidis (well developed), Hymenophyllum multifidum, Geranium sessiliflorum, Helichrysum bellidioides. On the Lyttelton Hills these species are not to be met with at the lower levels, though elsewhere on the peninsula, where the rainfall is higher, they come down much lower. Lycopodium fastigiatum, Veronica Lavaudiana, Gaultheria antipoda (the erect form), and Dracophyllum uniflorum also occur here in abundance. (8) The Forest. Distribution of the Forest Areas.—In pre-European times Banks Peninsula was clad with dense forest, except on the hills overlooking the plains and between Timutimu Head and Birdling's Flat, where the forests were chiefly confined to valleys and hilltops. From Lyttelton Heads to Kennedy's Bush there were no large forests, and between Kennedy's Bush and Little River the forests were not continuous. The hills on the southern side of Lyttelton Harbour were also in part bare; from Little River to Akaroa, and from Akaroa along the outer bays to Port Levy, there was an almost continuous sheet of forest. Here and there the bald heads of some of the higher peaks rose above the forest line, as in the case of Mount Herbert and Mounts Sinclair and Fitzgerald; but elsewhere the forest wave swept over the hilltops, and was continuous except where broken by cliffs. Certain trees abundant on the western ranges, which probably require a rainfall of 50 in., are conspicuously absent from the area: such are Metrosideros lucida, Weinmannia racemosa, and Phyllocladus alpinus. Types of Forest.—The forest is of two main types—(a) the podocarp forest, and (b) the beech forest.* This is not further described here, but see R. M. Laing, Trans. N.Z. Inst., vol. 46, p. 58, 1914. The podocarp forest can again be subdivided as to altitude into the lower podocarp and the upper podocarp-cedar

forest. It is characterized by the presence of certain taxads—viz., Podocarpus totara, Podocarpus spicatus, and in smaller quantity Podocarpus dacrydioides, as chief timber-trees. In the upper podocarp forest, on the tops of the ridges above 2,000 ft. P. totara is replaced by P. Hallii, P. dacrydioides is absent, and Libocedrus Bidwillii is found. This upper forest is found chiefly on the tops and sides of the ridges between Kaituna Valley and the Akaroa — Le Bon's dividing ridge. Here it passes into the beech forest. The Lower Podocarp Forest.—The lower podocarp forest is remarkably uniform throughout, and cannot be further subdivided by reference to the common species. The localization of some of the rarer, species suggests, however, a further subdivision into a seaward area lying between Port Levy and Akaroa, and a landward area lying between Lyttelton and Little River. The former of these areas is somewhat warmer and moister than the other. Here are to be found, though not uniformly distributed throughout the area, Corynocarpus laevigata, Rhopalostylis sapida, Coprosma lucida (large-leaved form), Cyathea medullaris, and a form of Sophora tetraptera approaching to but different from the East Cape S. grandiflora With the exception of Cyathea medullaris, these are plants which find their southernmost limit here. On the landward side these species are absent, but at one time Dacrydium cupressinum seems to have occurred. Certain rare species also occur here more commonly than on the eastern side. Thus both at Mount Pleasant and at Lake Forsyth there occur the following species, which do not reappear again together, so far as I know, elsewhere on the peninsula: Olearia fragrantissima, Teucridium parviflorum, Nothopanax anomalum, and Microlaena polynoda. Other localized species no doubt exist in the peninsula lowland forests, though it is difficult now to be sure that they were not at one time more widely distributed. Amongst these may be mentioned Pseudopanax ferox, now chiefly confined to the western area; Australina pusilla, in both areas; and Parietaria, in both areas. The distinctions of areas here made are not intended to be rigidly insisted on; they are, however, suggestive of some differences in primitive vegetation. But it is too late now to endeavour to define them more accurately, as we are not sufficiently well acquainted with the distribution of the species of sixty years ago. Certain species have undoubtedly disappeared since Raoul's time, and others have become very rare. These may be determined from the accompanying list. The taxads are the only large timber-trees common in the lower podocarp forest. Griselinia littoralis, however, often forms short, knotted trunks, sometimes 4 ft. through, but generally hollow. As the altitude increases, the forest-trees become somewhat smaller. Podocarpus spicatus, though not confined to the lower portion of the forest, is at its best and most abundant below 1,500 ft. Large trees of Podocarpus totara (sometimes in association with P. Hallii) are found up to 2,500 ft.; but in the upper parts of the “bush” Griselinia tends to replace the pines, and on the ridges and towards the summit Podocarpus Hallii and Libocedrus Bidwillii appear. The former is known on the peninsula as “mountaintotara,” and, though much valued as a timber, is considered inferior to P. totara. The white-pine is not, I think, found above 1,500 ft. Towards the upper limit of the forest the number of species is reduced and the specimens are stunted, until it finally gives place to a poorly developed subalpine scrub. The smaller trees of the forest are the same all over the peninsula to a remarkable extent, and, indeed, most of them are common as far south

as the Bluff. Griselinia littoralis, Melicylus ramiflorus, Carpodetus serratus, Pennantia corymbosa, Sophora microphylla, Nothopanax arboreum, Schefflera digitata, Aristotelia racemosa, Fuchsia excorticata, Pittosporum eugenioides, P. tenuifolium, and Hedycarya dentata are amongst the most abundant. These support an abundant growth of climbers, which in some places render the forest almost impenetrable. The most frequent are species of Rubus, Parsonsia, Clematis, and Muehlenbeckia; less common are Rhipogonum, Tetrapathaea, and Metrosideros hypericifolia. In the scrub which is found interspersed with the forest-trees are Drimys colorata, various species of Coprosma, Myrtus obcordata, Melicope simplex, and a few other plants. On the forest-floor are Blechnum discolor, and Polystichum vestitum in large quantities. By the side of the forest-streams Blechnum fluviatile, Blechnum lanceolatum, Asplenium bulbiferum, and Blechnum capense are the most abundant species of ferns. In the darker creeks Leptopteris hymenophylloides and Blechnum Patersoni appear. As a tree-parasite Loranthus micranthus is abundant, while Tupeia antarctica is much less common. There are, too, as in all parts of New Zealand, many epiphytic ferns: Cyclophorus serpens, Asplenium falcatum, Asplenium flaccidum, and Polypodium diversifolium are common in such situations. On exposed windy ridges the forest tends to pass out into the heath. The Upper Podocarp-Cedar Forest.—Through the centre of the peninsula, at an altitude of somewhat over 2,000 ft., runs a narrow belt of forest in which Podocarpus Hallii and Libocedrus Bidwillii are the predominant species. At its upward edge in places— e.g., Mounts Fitzgerald and Sinclair—this passes into a still narrower belt of subalpine scrub, probably not more than 100 or 200 yards in width, and not well defined. The absence of lowland trees, such as Myoporum, Dodonaea, and tree-ferns further differentiates this forest from that of the lowland area; otherwise it is very similar. Griselinia littoralis, Drimys colorata, Fuchsia excorticata, Nothopanax arboreum, and certain species of Coprosma are here in increased abundance, along with several species of Rubus. The Destruction of the Forest.—Unfortunately, however, there are only scraps of the forest left. Most of it has been destroyed with the advance of settlement. Possibly 10 per cent. of the big trees were used for timber, and the rest wastefully burnt. Of the extensive and almost continuous forest that once blanketed the peninsula from Little River round the coast to Pigeon Bay only scraps are left in the valleys and on the tops of the ridges, and practically everywhere these remnants are run through by cattle. This means that all the forest will be destroyed except for a few small and imperfect reserves. The destruction took place in the half - century between 1850 and 1900, and it is now practically complete. The largest area still forested is on the southern side of Mount Herbert. It is some two miles in length, and lies at an altitude of from 1,500 ft. to 2,500 ft. It is in too high and too bleak a situation to show the lower podocarp forest at its best, but it contains well-preserved areas of the podocarp-cedar forest and the subalpine scrub. Replacement of Vegetation in Forest Areas.—Where the bush land has not been sown with cocksfoot, but natural causes have been allowed to provide a secondary growth, considerable areas of bracken soon appear, and this in its turn is often replaced by heath, consisting chiefly of the two common species of manuka. Such areas are best seen where the bush was early destroyed by Maoris, settlers, or even by accident, as in Barry's Bay, Duvauchelle's Bay, Little River, Kaituna, and Port Levy.

Where the bush was partly burnt, or burnt patches occurred in the middle of the forest, fire weeds such as Aristotelia racemosa, Solanum aviculare, Erechtites prenanthoides, Plagianthus betulinus, and other fast-growing plants appeared, but, owing to the complete destruction of the forest, patches containing these second-growth species have now also disappeared. (9) The Lowland Scrub and Heath. Not all the scrub and heath areas are areas of second growth. There are scrub areas which have not at any recent time been covered with timber-trees, containing shrubs such as Melicope simplex, Corokia Cotoneaster, Helichrysum glomeratum, Coprosma crassifolia, C. areolata, C. virescens, Myrtus obcordata, Teucridium parviflorum, Myoporum laetum; and as climbers Clematis foetida, Rubus cissoides, Angelica geniculata, Parsonsia heterophylla var. rosea; whilst on the adjacent or interspersed rocks are Sophora prostrata, Libertia ixioides, and Clematis afoliata. Such scrub occurs where the rainfall has been too small or the exposure too great to admit of forest growth. All the species mentioned above occur on Mount Pleasant (behind Lyttelton) and again at Island Bay, and doubtless elsewhere. It constitutes a plant society which is characteristic of Banks Peninsula, and the same combination, so far as I know, does not occur elsewhere. Similar associations are to be found near the wharf at Port Levy, and also at Caton's Bay, though not so fully developed as in the two localities mentioned. With the scrub area I have associated the heath, which occurs at Governor's Bay, Port Levy, Duvauchelle's Bay, Barry's Bay, and Wainui, and is perhaps as well developed at Wainui as anywhere on the peninsula. Here it consists of Leptospermum ericoides (often covered with the parasite Viscum salicornioides), smaller quantities of L. scoparium, Gaultheria antipoda (the erect form), Cyathodes acerosa, Lycopodium volubile, Blechnum capense (a small form), Pteridium esculentum, Pteris scaberula, and the orchids Thelymitra longifolia and Pterostylis Banksii. Pseudopanax crassifolium occurs in it scantily, and elsewhere Leucopogon fasciculatum is a common constituent. Here also Raoulia subsericea appears exceptionally, and at an unusually low level of not more than 500 ft. A somewhat similar heath may be seen at the back of Mount Karetu. (10) The Subalpine Scrub. This consists, in addition to some species from the upper limits of the forest, of Olearia cymbifolia (forma), O. ilicifolia, with rare specimens of Aristotelia fruticosa and Gaultheria antipoda (the prostrate form). Veronica buxifolia was once found in it. This association is best seen on the old Purau line between Mounts Herbert and Fitzgerald, and again on the upper edge of the Mount Herbert bush. Species that might be expected and do not occur are Coprosma foetidissima, C. cuneata, and Veronica propinqua. Pentachondra pumila has been found on one or two hilltops. The poor development of the subalpine scrub may be due to an insufficient supply of moisture and a too high summer temperature. Near Dunedin it is much better developed, and includes two species of Dacrydium and Phyllocladus alpinus. These taxads are absent from Banks Peninsula. (11) The Subalpine Grasslands. Above 2,500 ft. there is to be found a distinct subalpine pasture.* See also Trans., N.Z. Inst., vol. 46, p. 58, 1914. It is characterized by the presence in abundance of Ourisia macrophylla

(forma), Drapetes Dieffenbachii, and Anisotome aromatica. Euphrasia zelandica, Forstera tenella, and Oreomyrrhis andicola also occur, but not so abundantly. This formation is found on all the tops from Mount Sinclair to Castle Rock, but for some reason does not occur on French Peak (2,600 ft.) to Saddle Mount (2,700 ft.). It reappears on Mount Bossu (2,500 ft.) and the neighbouring Carew's Peak (2,600 ft.), and traces of it reappear on Brasenose (2,500 ft.) behind Akaroa Township. Other species of the subalpine pasture are Danthonia Raoulii var. rigida, Aciphylla Colensoi, Raoulia glabra, and R. subsericea. Dantltonia Cunninghamii and Gunnera monoica are also sometimes to be found. This little subalpine florula is, of course, as much isolated as if it stood on an island off the coast, and it is somewhat difficult to account for its presence here. Either it has been brought by birds or winds, or else it is a remnant of a vegetation that in glacial times reached to the sea-coast and extended widely over the country. We have not the data at present to solve the problem, and it can only be considered in connection with the general geological evidence and with the distribution of alpine species throughout New Zealand. It is, of course, to be expected that in this latitude windswept mountain-tops should carry some subalpine species above 2,500 ft. Such species are found near Dimedin as low as 1,500 ft., and in Stewart Island at sea-level. In some cases, however (e.g., View Hill, French Peak), the forests were able to maintain themselves on hilltops up to 2,500 ft. This may be due to edaphic conditions, for these peaks are perhaps less rocky than those which are bare. Vertical Distribution. We have now considered briefly the general distribution of the indigenous vegetation on the peninsula. By way of brief recapitulation, it may be pointed out that in walking from the edge of Lake Ellesmere to the summit of Mount Herbert one would pass through practically all the plant associations described. Starting with the salt marsh and salt meadow, these follow in succession: the coastal scrub, a heath, a totara - black-pine forest, a Podocarpus Hallii - Libocedrus forest, and a narrow belt of subalpine scrub, and finally the subalpine pasture. The same series would be passed through in ascending from the beach at Pigeon Bay to the summit of Mount Sinclair, or from the shore of Lake Forsyth to the top of Mount Fitzgerald. Endemism. The area is so isolated that it might be expected to show a certain amount of endemism. This is most readily illustrated by the three species Celmisia Mackaui, Veronica Lavaudiana, and Senecio saxifragoides, Celmisia Mackaui has been reported from Mount Fyffe by H. B. Kirk, but the report has not been confirmed; and Veronica Lavaudiana has been recorded by Lyall, W. T. L. Travers, and J. B. Armstrong from the river-beds of the Canterbury Plains. Now, V. Lavaudiana is a true chasmophyte, and I should be very much surprised to see it growing in the shingle-bed of one of our Canterbury rivers. It might perhaps be expected in the mountain gorges of these rivers, but I have not seen it there, and much doubt its occurrence elsewhere than on Banks Peninsula. Travers records it from the Ashley, but though I have been at many points of the Ashley river-system I have seen nothing resembling V. Lavaudiana there. A form of V. Raoulii certainly occurs at White Rock in small quantity; and

V. Lavaudiana could scarcely, if occurring, have been completely overlooked by Cockayne, Wall, and myself. I must therefore consider the above three species as endemic on the peninsula till more definite evidence is brought to the contrary. Cotula Haastii is another species which may be endemic on the peninsula; though recorded from the plains by Haast, it has not been found there recently. There are other plants which are here represented by varietal forms—e.g., Aciphylla squarrosa, Angelica montana, Anisotome Enysii, Myosotis australis, Ourisia macrophylla, Veronica Lyallii, &c.; but as it is not my intention to deal with critical species in this paper I am leaving these out of consideration here. Enough has been said to show that there is sufficient endemism in the area to suggest its isolation for a long time from any other portion of the country. Species at their Southernmost Limit. The following eighteen species probably here reach their southernmost limit on the eastern coast. Some are to be found farther south on the west coast. I give, where possible, some notes on their northward distribution. Alectryon excelsum.—Abundant to the north of the Hurunui, and probably to be found in coastal valleys as far south as the mouth of the Waipara (Double Corner). Angelica rosaefolia.—If correctly reported by Kaoul from Akaroa this shows a remarkably discontinuous distribution, as it is not otherwise known outside the North Island. Clematis Colensoi.—Found also on the foothills to the west; ocal on the peninsula. Cordyline indivisa.—Does not occur on the coastal hills of North Canterbury, and I have seen no record of it from the Kaikouras, where, however, it may be expected to occur. Corynocarpus laevigata.—Common on the Kaikoura coast and as far south as Gore Bay; a single tree at Manuka Bay, north of the Hurunui; but does not, I think, occur between the mouth of the Hurunui and Banks Peninsula. Cyathea Cunninghamii. — If the determination be correct this is a remarkable southern extension of its range on the east of the South Island, but in Westland it extends beyond the Big Wanganui River. Dodonaea viscosa.—Common along the coast to the north of Pegasus Bay. Griselinia lucida.—I have not seen this south of Banks Peninsula, and doubt its occurrence, though reported in the Manual as not uncommon as far south as the Bluff. To the north of Banks Peninsula I have noted its occurrence at Manuka Bay, where typical specimens are to be seen on rocks, just to the north of the mouth of the Hurunui. It is not reported in the Dunedin Field Club's Catalogue (1916). Hedycarya arborea.—On the Kaikoura coast, but I have not seen it between the Amuri Bluff and Banks Peninsula. It is curious that on the west coast it should occur as far south as Preservation Inlet, but it does not pass beyond Banks Peninsula here. Possibly the arid Canterbury Plains have proved a barrier to its southward progress. Leucopogon fasciculatum.—Occurs inland on Mount Grey downs and in the Lee Valley.

Mariscus ustulatus.—Occurs at Gore Bay. It is reported from North Otago by Buchanan, but this requires confirmation. Macropiper excelsum.—Common on the coastal hills to the north, but not on the plains. Pittosporum obcordatum.—If Raoul's Akaroa record is correct this disappearing species shows remarkable discontinuity of distribution. Only known elsewhere from near Kaitaia. Rhopalostylis sapida.—Does not occur on the coastal hills of North Canterbury, but is found on the Kaikoura coast as far south at least as Hundalee. Rhagodia nutans.—Not uncommon near the coast at least as far south as Banks Peninsula. Spiranthes australis.—Not known elsewhere on the east coast of this Island. Tetrapathaea australis.—I do not know the nearest point where this plant is to be found north of Banks Peninsula, but it does not occur in northern Canterbury, nor, I think, at Kaikoura. Zoysia pungens is found on the beaches immediately to the north, and shows no discontinuity of distribution. It thus appears that Banks Peninsula is the southern terminus of a large number of northern species. On the other hand, few if any southern species find here their northern limit; Olearia fiagrantissima is perhaps one. This suggests a southern drift of species; for, of course, most of the northern species could live as well on the Otago coast as in Banks Peninsula. On the west coast, where there has been no such barrier as the long treeless stretch between Banks Peninsula and Timaru, some of these species have drifted much farther south—e.g., Cordyline indivisa, Hedycarya arborea, and Rhopalostylis sapida. Obviously the frosts of the plains tend to inhibit the occurrence of the more tender species. The Relationships of the Banks Peninsula Florula. The Forest.—It will be clear from what has been said that the forest of Banks Peninsula is an outlier of Cockayne's North-eastern Botanical District, No. 8 in the map of New Zealand* See also Trans., N.Z. Inst., vol. 46, p. 58, 1914. showing proposed botanical districts. It does not bear any close relationship to the forests of the Canterbury foothills. The nearest hills are those to the north and north-north-west—e.g., Mounts Oxford, Karetu, and Grey. Here all the species finding their southernmost limit on Banks Peninsula are absent, with the exception of Clematis Colensoi and Leucopogon fasciculatum; and the timbertrees belong chiefly to the genus Nothofagus. It is true that Podocarpus totara is present in places on the foothills in small quantity, and P. spicatus and P. dacrydioides in still smaller quantities, but they are nowhere, as on the peninsula, the dominant species. We get, however, to the westward at Mount Peel a mixed podocarp forest in which the above-mentioned species predominate, but it, too, differs much from the Banks Peninsula forest. This, of course, does not contain the species finding their southernmost limit on Banks Peninsula, but it contains an unusual variety of species, and amongst them several not found on the peninsula, such as Nothopanax simplex, Aristotelia Colensoi, Gaya Lyallii var. ribifolia, and Hoheria lanceolata. It thus appears that the Banks Peninsula forest must be regarded as an outlier of the Kaikoura coastal forest, which it closely resembles.

The Subalpine Area.—This area, as might be expected, is related to the subalpine area of Mounts Grey and Karetu, hills of about the same height as Mount Herbert, and little more than thirty miles away to the north. In both localities Drapetes Dieffenbachii, Anisotome aromatica, and Olearia cymbifolia are amongst the first subalpine forms to appear above the forest line. Mount Oxford is higher and has a much more varied alpine vegetation; and even Mount Karetu from 1,500 ft. and upwards shows species not occurring on the peninsula: Amongst them are Coprosma repens, Celmisia spectabilis, Anisotome filifolium. Exocarpus Bidwillii, and Senecio geminatus. No doubt this is due to the fact that we have here to deal with a much larger subalpine area than on the peninsula. So far as I know, however, Ourisia macrophylla, Forstera tenella, and Oreomyrrhis andicola do not occur in the Mount Grey district. If further search does not reveal their presence, their absence might seem to indicate that the subalpine florula of the peninsula contains species belonging to an older alpine flora, not now existent in the neighbourhood. None of the plants referred to are species likely to have been recently brought there, or to have been brought in the first place by wind or birds. Still, we have not enough evidence here to come to any definite conclusions, and the matter must be left in abeyance. This review of the vegetation of Banks Peninsula must now be concluded, with the hope that before the present remnants of the primitive flora disappear every opportunity will be taken by local students to complete the work here outlined. Before concluding this section I must thank those New Zealand botanists from whom I have received much kind assistance. Mr. Cheeseman has identified a number of species for me, particularly of ferns; as also has Mr. Petrie, who has given me much help in connection with the genera Coprosma and Uncinia. I owe much to the kindly suggestions and advice of Dr. Cockayne, who has studied the vegetation of the Port Hills; and Professor Wall, who has during several years closely examined the flora of the peninsula, has kindly revised the distribution of the species, and, as will be seen, made numerous additions to the list of localities given. I am particularly indebted to him for assistance with the ferns and with the genus Epilobium. I believe that the list as it now stands is fairly complete, and that the number of subsequent additions will be comparatively small. Bibliographical List of Chief Books and Articles consulted. 1846. A. Raoul, Choix de plantes de la Nouvelle-Zélande. 1853. Sir J. D. Hooker, flora Novae-Zelandiae. 1864. ——, Handbook of the New Zealand Flora. 1870. J. F. Armstrong, On the Vegetation of the Neighbourhood of Christchurch, Trans. N.Z. Inst., vol. 2, p. 118. 1880. J. B. Armstrong, A Short Sketch of the Flora of the Province of Canterbury, Trans. N.Z. Inst., vol. 12, p. 325. 1882. T. H. Potts, Out in the Open. 1899. T. Kirk, Students' Flora. 1910. Miss B. D. Cross, Observations on some New Zealand Halophytes, Trans. N.Z. Inst., vol. 42, p. 545. 1914. R. M. Laing, On a Subalpine Element in the Flora of Banks Peninsula, Trans. N.Z. Inst., vol. 46, p. 56.

1914. Miss L. A. Suckling, The Leaf-anatomy of some Trees and Shrubs growing on the Port Hills, Christchurch, Trans. N.Z. Inst. vol. 46, p. 178. 1915. L. Cockayne, Provisional List of Ferns and Flowering-plants of the Port Hills, Report on Scenery-preservation (1914-15). 1918. A. Wall, Ferns of the Port Hills, Lyttelton Times, 13th July, 1918. 1918. ——, On the Distribution of Senecio saxifragoides Hook. f., and its Relation to Senecio lagopus Raoul, Trans. N.Z. Inst., vol. 50, p. 198. List of Indigenous Plants Found on Banks Peninsula. Explanation of Abbreviations and Signs. 1. Plants marked thus ° (e.g., °Melicytus micranthus) in the list are those that I have not collected myself or seen in the collections of others, but are introduced on the evidence of previous collectors. It may perhaps be noted that Raoul's list has been very carefully drawn up, and is throughout reliable. Though the identifications of Mr. J. F. Armstrong's list may not always be correct, still some plant can generally be found which the name represents; but in J. B. Armstrong's list there are names of many species which obviously do not occur on Banks Peninsula. Some of these can at once be rejected on external evidence, but in other cases it is manifestly impossible to say that the plant has not been found on the peninsula, though its occurrence there may be highly improbable. It is included in my list if there is any subsidiary evidence to suggest that it may have become extinct or have been overlooked. These species inquirendae are marked with a small (1) before the initial letter of the genus—e.g., (1)Eleocharis Cunninghamii. This is intended to indicate that there is some reason to believe that the plant occurs or has occurred on the peninsula, but that it has not been recently found. Similarly the species excludendae—those plants which, though recorded, probably have not been found on the peninsula, or which have been identified in error—are marked with a small (2) before the initial letter of the genus — e.g., (2)Fimbristylis fiondosa. Of course, the line between species inquirendae and species excludendae is often very indistinct. It is quite possible that some of the species excludendae may subsequently be found on the peninsula, but the evidence in their favour does not justify their inclusion at present. Should they be discovered they can be readily reinstated. Of the species inquirendae, some are included in my list, and others, where perhaps the evidence for their occurrence is somewhat weaker, are only noted. In any count of the list, of course, only those which are definitely included should be reckoned. 2. Plants whose names are preceded by an asterisk, thus, *Asperella gracilis, have not been found on the Lyttelton Hills—i.e., between Gebbie's Pass and Lyttelton North Head. Plants not so marked are to be found or have in the past been found on the Lyttelton Hills. Some are perhaps now extinct there. It will also be noted that some plants recorded from the Lyttelton Hills are not recorded from Banks Peninsula; this in the majority of cases is probably due to the fact that the district nearer Christchurch has been more exhaustively examined than the more remote one. This, however, is not always the case, for Senecio saxifragoides and Myosotis australis var. seem to be confined to the smaller district.

3. Initials as under, which will be readily recognized, are used to indicate records where no details are given. In many cases such records will appear under synonyms and not under the names now in use. R. M. E. Raoul. J. F. A. J. F. Armstrong. J. B. A. J. B. Armstrong. T. P. T. H. Potts. L. C. L. Cockayne. A. W. Arnold Wall. T. K. Thomas Kirk. Where no authority is given for a locality it will be understood that it is recorded from my own observations. Occasionally, to avoid confusion, I have introduced my own initials (R. M. L.). By “Handbook” is meant Hooker's Handbook of the New Zealand Flora, and by “Manual” Cheeseman's Manual of the New Zealand Flora. 4. No doubt there are many localities still to be found for some species. Some plants may be comparatively common which are recorded from only one or two places. Where it is not definitely known that the plant is widespread it is probably better to give exact habitats than make a vague statement. The list will, I think, be found practically complete as regards the Lyttelton Hills, but there will doubtless be a few omissions to be filled in for the peninsula, particularly in the Cyperaceae and grasses. A good many varieties and a few critical species require closer description and identification. It would, however, much increase the length of this paper to deal with them, and it would, further, require comparison with forms in other parts of New Zealand. This work lies outside of the scope of the present article, but it may be hoped that this discussion will be carried on subsequently by myself or some other botanist. Any collector of Banks Peninsula plants should have no difficulty in identifying the plant referred to in my list. References will be found in the list to plants requiring further investigation. Ferns (Filices). Hymenophyllum rarum R. Br. [J. B. A.] Waikerikikeri: R. M. L. Rapaki: Potts. “Near Christchurch some of the dwarfer forms may be observed near the top of the small patches of bush that dot the Native reserve at Rapaki” (Out in the Open, p. 61). Hymenophyllum sanguinolentum Hook. f. Mount Pleasant: A. Wall! Banks Peninsula: Potts; J. B. A. Hymenophyllum villosum Col. Castle Rock (Con. Cheeseman). *°(1)Hymenophyllum australe Willd. Banks Peninsula: Potts. °(1)Hymenophyllum dilatatum Swartz. Lyttelton Hills: Potts. Banks Peninsula: J. B. A. I have searched the most secluded gullies on the peninsula in vain for this and the next species. °(1)Hymenophyllum demissum Swartz. Banks Peninsula: Potts; J. B. A.

*Hymenophyllum flabellatum Labill. [J. B. A.] Mount Herbert Peak, Peraki Reserve: R. M. L. *°Hymenophyllum Malingii Metten. [J. B. A.] “On Libocedrus Doniana [i.e., L. Bidwillii], Port Levy”; Potts (Out in the Open, p. 73). °(1)Hymenophyllum minimum A. Rich. Rapaki: Potts. Banks Peninsula: J. B. A. Hymenophyllum tunbridgense Smith. [J. B. A.; L. C.] Lyttelton Hills: Potts. Charteris Bay: A. Wall. Dover Castle and Mount Fitzgerald: R. M. L. *Hymenophyllum unilaterale Willd. Long Bay, behind Akaroa; apparently a new record. Hymenophyllum multifidum Swartz. [J. B. A.] Cilff near Cooper's Knobs; Castle Rock, and Mount Fitzgerald above 2,000 ft., on rocky southern faces. *°(1Hymenophyllum bivalve Swartz. Banks Peninsula: Potts; J. B. A. Some of the above species are probably now extinct on the peninsula. J. B. A. records also (2)H. scabrum, (2)H. pulcherrimum, and (2)H. subtilissimum. *Trichomanes venosum R. Br. [R.; T. P.; J. B. A.; L. C.] Balgueri Valley (Akaroa), on tree-fern trunks: R. M. L. J. B. A. records also (2)T. humile, (2)T. elongatum, (2)T. (Hymeno-phyllum) Lyallii, and (2)T. Colensoi; but their existence on the peninsula is unconfirmed. Cyathea dealbata Swartz. [R.; J. F. A.; J. B. A.; T. P.; A. W.; L. C.] Once abundant; now becoming less common. Tree-ferns have been largely used for Christmas and other decorations in Christchurch, and now are almost extinct on the Lyttelton Hills, where formerly they abounded. *Cyathea medullaris Swartz. Apparently very rare on Banks Peninsula, but a few plants occur in the remnants of the bush at Waikerikikeri: R. M. L. Akaroa: Raoul Cyathea Cunninghamii Hook. f. (?). Bush near Cooper's Knobs: A. Wall! The identification is not quite certain as yet; but the fern discovered can scarcely be anything else, though specimens sent to Mr. T. F. Cheeseman are pronounced to be probably Hemitelia Smithii. Hemitelia Smithii Hook. f. [J. B. A.; T. P.; A. W.] Akaroa; Wainui; Kaituna Valley: Abundant in the last two places, and probably elsewhere. It just crosses Gebbie's Pass to the bush just below Cooper's Knobs. Dicksonia squarrosa Swartz. [R.; J. F. A.; J. B. A.; T. P.; L. C.; A. W.] Common on the peninsula, but almost extinct on the Lyttelton Hills. *Dicksonia fibrosa Col. (?). Wainui. As I have no specimens I am somewhat doubtful of the identification. (1)D. lanata is also recorded by J. B. A. and J. F. A., and (1)Alsophila Colensoi by J. B. A. and T. P.

*Davallia novae-zealandiae Col. [J. B. A.; T. P.] Banks Peninsula: William Martin (Lyttelton Times, 17th August, 1918). I believe I collected this species some thirty years ago at Pigeon Bay, but have not seen it recently. (1)Cystopteris fragilis is recorded by J. B. A. and T. P. Adiantum affine Willd. [R.; J. F. A.; J. B. A.; A. W.] Becoming rarer, but still growing on the southern faces of the Lyttelton Hills, Governor's Bay; in abundance at Nikau Palm Gully (Akaroa) and Caton's Bay. *Adiantum fulvum Raoul. Akaroa: Raoul. Charteris Bay: R. M. L. Evidently by no means common. *°(1)Adiantum aethiopicum Linn. [J. B. A.] Akaroa: Raoul. This has not been seen in recent times, and must be regarded as a doubtful species. (2)A. hispidulum has also been reported, but I have not seen it. Hypolepis tenuifolia Bernh. [J. F. A.; J. B. A.; L. C.] Ohinitahi: Potts. Common, usually among stones or boulders above the bush line; perhaps extinct on the Lyttelton Hills. Hypolepis millefolium Hook. [J. B. A.; T. P.] At one time on the Lyttelton Hills above Kaituna, and no doubt elsewhere. Hypolepis distans is also reported by J. F. A. and J. B. A. Cheilanthes Sieberi Kunze. [R.; J. F. A.; T. P.; J. B. A.; L. C.; A. W.] Abundant in rocky clefts, and under stones, in dry situations. Cheilanthes tenuifolia Swartz. [R.; J. B. A.; T. P.; T. K.; L. C.; A. W. Lyttelton Hills and Banks Peninsula, in tussock lands amongst rocks; now very rare. Pellaea rotundifolia Hook. [R.; J. F. A.; J. B. A.; L. C.; A. W.] Common in forest and by streams; occasionally under rocks in dry places. Pteridium esculentum Cockayne. [R.; J. F. A.; J. B. A.; L. C.; A. W.] Everywhere abundant; often coming up after the burning of the forest. *Paesia scaberula Kuhn. [R.; J. F. A.; J. B. A.] Not common; Mount Bossu, Wainui: R. M. L. Histiopteris incisa J. Sm. [J. F. A.; J. B. A.; L. C.] Ohinitahi: Potts. Not common; on the damper hilltops, Pigeon Bay, Lyttelton Hills (probably extinct), Stony Bay. *°(1)Pteris tremula R. Br. [J. B. A.] Near Tikao Bay: Potts. Cheeseman. (Manual, p. 793) mentions (2)P. macilenta as recorded from Banks Peninsula, but has seen no specimens. *Blechnum† The species of Blechnum appear in the Manual under the generic name Lomaria. Patersoni Mett. [J. B. A.; T. P.] Common in damper forests.

Blechnum discolor Keys. [R.; J. F. A.; J. B. A.; L. C.; A. W.] Common on the forest-floor. Blechnum lanceolatum Sturm. [R.; J. F. A.; L. C.; A. W.] Common in the forest. Blechnum penna marinum Kuhn. [R.; J. F. A.; J. B. A.; T. P.] Abundant above 1,000 ft., but not common on the Lyttelton Hills, and coming down to 300 ft. in Pigeon Bay. Heathcote Valley: A. Wall; R. M. L. Peak between Kennedy's Bush and Cooper's Knobs: R. M. L. Blechnum capense Schlecht. [R.; J. F. A.; J. B. A.; T. P.; L. C.; A. W.] Everywhere abundant. Blechnum fluvatile Lowe [J. F. A.; J. B. A.; L. C.; A. W.] Common by bush-streams. Blechnum membranaceum Mett, [J. B. A.; T. P.; L. C.; A. W.] Hay's Bush (Pigeon Bay) and behind Governor's Bay, but not common. (1)B. Banksii, (2)B nigrum, and (1)B. durum are given by J. B. A. (the last probably occurs), and (1)B. vulcanicum by J. F. A. and T. P. (Ohinitahi). Cheeseman (Illustrations, pl. 240) refers to (1)B. durum finding its northernmost limit at Banks Peninsula. Asplenium flabellifolium Cav. (= ? A. triste Raoul). [R.; J. B. A.; J. F. A.: T. P.; L. C.; A. W.] Abundant under stones and rocks. *Asplenium adiantoides Raoul. [J. B. A.; T. P.] Common on peninsula. Asplenium obtusatum Forst. f. [R.; T. P.; J. B. A.; L. C.] Abundant on coastal cliffs and rocks; but it is to be noted that Wall identifies the coastal Asplenium of Lyttelton Harbour as a form of A. lucidum. Asplenium lucidum Forst. f. [R.; J. F. A.; J. B. A.; A. W.] Common in most places. Var. obliquum is the commonest, and perhaps the type form, but var. Lyallii also occurs. [Con. L. C.] Asplenium Hookerianum Col. [R.; J. F. A.; J. B. A.; T. P.; A. W.; L. C.] Common in many forms. (Akaroa is the habitat of the type.) Asplenium bulbiferum Forst. f. [R.; J. F. A.; J. B. A.; L. C.; A. W.] One of the commonest of bush ferns, but much sought after by tourists and others. *Asplenium Richardi Hook. f. [R.; J. B. A.] Mount Herbert: Potts. Waikerikikeri, and doubtless elsewhere: R. M. L. Asplenium flaccidum Forst. f. [R.; J. F. A.; J. B. A.; T. P.; L. C.; A. W.] Abundant on forest-trees. J. B. A. and Potts add (1)A. trichomanes, which I have not seen. Polystichum vestitum Presl. (= Aspidium aculeatum Swartz var. vestitum Hook. f.: Handbook, p. 997). [R.; J. F. A.; J. B. A.; T. P.; L. C.; A. W.] Common on the forest-floor.

Polystichum Richardi J. Sm. (= Aspidium Richardi Hook. f.: Handbook, p. 999). [J B. A.; T. P.; L C.; A. W.] Akaroa: Raoul. Abundant in open tussock pasture. This includes doubtless the varietal form P. oculatum Hook. f. Polystichum hispidum J. Sm. [J. B. A.; T. P.; L. C.] Otahuna: A. Wall! *°(1)Polystichum capense J. Sm. Akaroa: Raoul. Probably an erroneous identification. Dryopteris glabella C. Chr. [J. B. A.; L. C.] Allandale: A. Wall! Balgueri Valley: R. M. L. Ohinitahi: Potts, Dryopteris punctata C. Chr. [J. B. A.; T. P.; L. C.; A. W.] Common. Dryopteris pennigera C. Chr. (= Polypodium pennigerum Forst. f.: Manual, p. 1009). [J. B. A.; T. P.; L. C.; A. W.] Common by the sides of streams. J. F. A., J. B. A., and T. P. give also (1)Dryopteris (Nephrodium) decomposita *°(1)Nephrodium velutinum Raoul. [J. F. A.; J. B. A.; T. P.] Akaroa. *Polypodium australe Mett. [J. B. A.; T. P.] Common on trees and rocks. Polypodium australe Mett. var. pumilum Cheesm. Mount Herbert Peak, and doubtless elsewhere; near Cass Peak: R. M. L. *(1)Polypodium Cunninghamii Hook. [J. B. A.] Akaroa: Raoul. A rather doubtful inhabitant of the peninsula. †Polypodium pustulatum Forst. f. [J. B. A.] Akaroa: Raoul. I was somewhat surprised to find in Caton's Bay (Lake Forsyth), just above the house, on the forest-floor, undoubted specimens of this species, as I had considered it either extinct on the peninsula or wrongly identified. Polypodium grammitidis R. Br. [R.; J. F. A.; J. B. A.; T. P.; A. W.] Not uncommon on rocks exposed to the south-west. Polypodium diversifolium Willd. [R.; J. F. A.; J. B. A.; T. P.; L. C.; A. W.] Everywhere abundant on rock and forest-tree. J. B. A. gives also (1)Arthropteris (Polypodium) tenella. Cyclophorus serpens C. Chr. [R.; J. F. A.; J. B. A.; A. W.] Cliffs, rocks, and trees; common. Nothoclaena distans R. Br. [J. B. A.; T. P.; L. C.; A. W.] Lyttelton Hills, Quail Island, and Lyttelton Harbour generally: A. Wall! An undescribed Cheilanthes or Nothoclaena is common about Diamond Harbour, where it has been found by A. Wall. Cheeseman considers it to be probably a Nothoclaena. Gymnogramme rutaefolia Hook. & Grev. [L. C.; A. W.] Dry rocks, Dover Castle, and north side of Mount Pleasant; also Sumner Valley; Wakeroa Head; Pigeon Bay.

Gymnogramme leptophylla Desv. [J. F. A.; J. B. A.; T. P.; L. C.; A. W.] In open tussock land; not common. Heathcote Valley: E. Holds-worth! Lyttelton. *Gleichenia Cunninghamii Heward. [J. B. A.; T. P.] Port Levy, about 2,000 ft., on Pigeon Bay side of valley, but now extinct in this locality. Recorded by William Martin behind Le Bon's Bay (Lyttelton Times, 17th July, 1918). Potts records also (2)Gleichenia dicarpa, and J. B. A. (2)G. circinata. It is unlikely that either occurs. Leptopteris hymenophylloides Presl. [J. F. A.; J. B. A.; T. P.; L. C.; A. W.] In the darker forests, Cooper's Knobs, Kaituna, Port Levy, &c. J. B. A. gives also (2)Todea superba and (2)Schizaea dichotoma; both are unlikely species. Ophioglossum coriaceum A. Cunn. [J. B. A.; L. C.] Lyttelton Hills, near Bridle-path, hills behind Wainui; ridge between Le Bon's and Waikerikikeri; and elsewhere. Botrychium australe R. Br. [J. F. A.; J. B. A.; T. P.; L. C.] Mount Herbert; Lyttelton Hills, &c.; but not often found now. (2)Phylloglossum Drummondii is reported by J. B. A., but probably in error. Lycopodium varium R. Br. Lyttelton Hills (perhaps now extinct); Governor's Bay, Caton's Bay, Waikerikikeri, and in several other localities, but not now common. *Lycopodium fastigiatum R. Br. [R.] Common on the mountain-tops in the centre of the peninsula; also on peak between Cass Peak and Cooper's Knobs. Lycopodium volubile Forst. f. [J. B. A.] Akaroa: Raoul. Wainui hilltops, above Le Bon's Bay; Summit Road, near Hilltop Hotel: R. M. L. Mount Herbert: A. W. J. B. A. gives also (2)L. Selago and (2)L Billardieri. Lycopodium scariosum Forst. f. Only noted in the patch of forest beyond Kennedy's Bush. *Tmesipteris tannensis Bernh. [J. B. A.] Akaroa: Raoul. Okute Valley, Little River: R. M. L. Class CONIFERAE. Family Taxaceae. *Libocedrus Bidwillii Hook. f. [J. B. A.] Akaroa to Mount Herbert, above 1,500 ft., and chiefly on the southern sides. (2)L. Doniana, recorded by J. B. A., does not occur. Podocarpus totara D. Don. [J. F. A.; J. B. A.; L. C.] At one time everywhere abundant; now getting uncommon. *Podocarpus ferrugineus D. Don. [J. F. A.; J. B. A.] At one time there were a few trees on the peninsula. The only ones I have seen recently are in a small group in Port Levy, on the slopes of Mount Herbert, at about 1,200 ft. Mr. E. Hay has informed me that there is one specimen in his reserve at Pigeon Bay.

*Podocarpus Hallii T. Kirk. Usually towards the tops of the hills, where it is often rather stunted. Generally known on the peninsula as “mountain-totara.” The timber is considered somewhat inferior to that of P. totara. Apparently not hitherto recorded. Podocarpus spicatus R. Br. [J. B. A.; L. C.] With P. totara this was at one time the chief timber-tree of the forests, now largely destroyed. Above 2,500 ft. it becomes dwarfed and flattened to the rocks. Podocarpus dacrydioides A. Rich. [J. B. A.] Much less common than the preceding, and generally on the flats, though scattered trees were found in most valleys. “The flat was covered [i.e., in Le Bon's Bay] with white and black pines as thick as they could stand; and the sides of the valley grew immense totaras and other timber.” (H. C. Jacobson, Tales of Banks Peninsula, 1893.) *(1)Dacrydium cupressinum Soland. [J. B. A.] I introduce this with hesitation, as I have not seen it on the peninsula, and believe it to be now extinct. However, settlers have assured me that it once grew in Little River, near Okain's, and elsewhere in small quantity. Mr. W. H. Montgomery, of Little River, writes thus: “Rimus used to be common on my land near the Hilltop”; and I have heard from other sources that some 40,000 ft. of this timber was taken from a stand on Harman's Track (Puaha). J. B. A. (Trans., vol. 12, p. 328) states the rimu is “chiefly found on the higher ridges, and is here a far inferior tree in beauty compared to the West Coast variety of the same species.” This does not agree with my information; and I certainly would expect to find it rather in the moister valleys amongst the denser bush than on the higher ridges. Class MONOCOTYLEDONEAE. Family Typhaceae. Typha angustifolia Linn. [J. F. A.; J. B. A.] In swampy places common, though becoming less abundant as drainage proceeds. Family Naiadaceae. Triglochin striatum Ruiz. & Pav. [J. B. A.] Kaituna; Heathcote Estuary: R. M. L. Children's Bay, Akaroa: A. W. *Ruppia maritima Linn. [J. B. A.] Kaituna, Lake Ellesmere. Zostera nana Roth(?). [J. B. A.] I have not examined this species carefully. Lyttelton Harbour and elsewhere. *°Potamogeton pectinatus Linn. Lake Forsyth: T. Kirk. *°Potamogeton ochreatus Raoul. [R.; J. B. A.] Banks Penmsula is the habitat of the type. J. B. A. records two other species—(2)P. natans and (2)P. compressus—but I have not seen them. P. Cheesemanii is to be expected, but has not yet been recorded.

Family Gramineae. °(1)Zoysia pungens Willd. [J. B. A.] Recorded by Cheeseman (Manual, p. 844). I have seen it growing at Gore Bay, and have also seen specimens from New Brighton, and consider that it probably still occurs on the peninsula. Microlaena avenacea Hook. f. [R.; J. B. A.; L. C.] Abundant on the forest-floor. (Akaroa is the habitat of the type.) Microlaena polynoda Hook. f. [J. F. A.; J. B. A.; L. C.] Rare. Redcliffs, Caton's Bay, in scrub; bush beyond Kennedy's: A. W. Hierochloe redolens R. Br. [J. F. A.; J. B. A.] Common; usually above 800 ft. *Stipa arundinacea Benth. [J. F. A.; J. B. A.] Pigeon Bay; Long Bay (near Peraki): R. M. L. Akaroa: Lyall; Kirk. Stipa setacea R. Br. Sumner; Heathcote Valley; Cashmere Hills. Probably a recent introduction. Echinopogon ovatus Beauv. [J. B. A.] Apparently not common. Lyttelton Hills: A. W. Caton's Bay; Island Bay: R. M. L. J. B. A. records (1)Agrostis canina and (1)A. quadriseta. These should certainly be looked for. *Deyeuxia filiformis Petrie. [J. B. A.] Kaituna; Mount Fitzgerald; Stony Bay. *°Deyeuxia filiformis Petrie var. pilosa Cheesem. Banks Peninsula: T. Kirk. Dichelachne crinita Hook. f. [J. B. A.; L. C.] Common on rock-faces and drier hillsides Deschampsia caespitosa Beauv. [J. F. A.; J. B. A.] Akaroa: Raoul. Tops of ridges, Lyttelton Hills: R. M. L. And probably elsewhere. °Trisetum antarcticum Trin. [J. B. A.] Lytteelton Hills: Cockayne. *Danthonia Cunninghamii Hook. f. Mount Herbert and Mount Sinclair, 2,500 ft. Apparently not hitherto recorded. Danthonia Raoulii Steud. Akaroa (the habitat of the type); Mount Sinclair, above 1,000 ft.; Cooper's Knobs. I treated this previously as D. flavescens (Trans., vol. 46, p. 58), but as the plant differs somewhat from the form as found elsewhere, and as it is undoubtedly the D. rigida of Raoul, it is the type of D. Raoulii and may be called var. rigida. Danthonia semiannularis R. Br. [R.; J. B. A.; L. C.] Common in the tussock meadows, and probably increasing in quantity. Danthonia pilosa R. Br. [L. C.] Abundant, and increasing (see introduction, p. 363).

Arundo conspicua Forst. f. [J. F. A.; J. B. A.; L. C.] Common in gullies with flax. Poa caespitosa, Forst. f. [J. B. A.; L. C.] The chief tussock of the hillside, but slowly disappearing owing to fires and cultivation. Poa Colensoi Hook. f. [J. B. A.; L. C.] Commoner towards the hilltops, and apparently a disappearing species. Poa imbecilla Forst. f. var. Matthewsii Hack. [J. B. A.; L. C.] Common in damper and shadier ground near the hilltops. J. B. A. includes also (2)P. breviglumis, (2) P. foliosa, (2)P. anceps, (2)P. Lindsayi, and (2)P. scoparia. Two of these are plants of the subantarctic islands. Atropis stricta Hack. var. suborbicularis Hack. Salt meadows, Teddington (det. Petrie). Apparently not hitherto recorded. Festuca novae-zealandiae Cockayne. [J. B. A.; L c.] Not uncommon in the tussock pastures. °Festuca multinodis Petrie. Lyttelton Hills: Cockayne. *Festuca littoralis Labill. [J. B. A.] Peninsula, near shore. Agropyron scabrum Beav. var. [L. C.] Lyttelton Hills; not common. Island Bay: R. M. L. A migrant tussock. *Agropyron multiflorum T. Kirk. Akaroa: T. Kirk. (Specimen in the Canterbury Museum.) *°Asperella gracilis T. Kirk. [J. B. A.] Akaroa: Raoul. Family Cyperaceae. The following species, recorded only by J. B. A., should, I think, be struck off the list: (2)Cyperus tenellus, (2)Schoenus tenax, (2)S. pauciflorus, (2)S. axillaris. *Mariscus ustulatus C. B. Clarke. [J. B. A.] Peraki: L. C. Near Long Lookout, Island Bay: R. M. L. Eleocharis acuta R. Br. [J. B. A.] In damp places, Cashmere Hills: R. M. L.; A. W. (1)Eleocharis Cunninghamii Boeck. [J. F. A.; J. B. A.] Lyttelton Hills: L. C. (Possibly in error for the preceding species, though A. W. has found it in Hagley Park.) (1)E. sphacelata is also recorded by J. B. A. Scirpus inundatus Poir. [J. B. A.] Damp spots on Lyttelton Hills: A. W. Doubtless also on Banks Peninsula. *°Scirpus antarcticus Linn. Mount Herbert: A. Wall.

*Scirpus sulcatus Thouars var. distigmatosa C. B. Clarke. Wainui, in watercourses. Scirpus nodosus Rottb. [J. B. A.] Lyttelton Hills; Pigeon Bay; Akaroa. Common in dry watercourses and near the seashore. Scirpus cernuus Vahl. Salt meadows, Teddington, and elsewhere. Scirpus frondosus Banks & Sol. [J. F. A.] Sandhills, Sumner and Taylor's Mistake. Scirpus americanus Pers. Ohinitahi; Lake Ellesmere; &c. Scirpus lacustris Linn. Heathcote: B. D. Cross. Scirpus maritimus Linn. Akaroa: T. Kirk. Lake Ellesmere; Ohinitahi. J. B. A. gives also (2)Fimbristylis dichotoma, a very improbable inhabitant, being a plant of the warmer portions of the North Island, but (1)Cladium teretifolium may possibly occur. Four species of Gahnia also appear on his list. I have seen none, and if any occur they are by no means common. J. F. A records (2)Lepidosperma tetragona (= Cladium Vauthiera): this may possibly occur, but I have not seen it nearer than Mount Grey. Cladium glomeratum R. Br. [J. B. A.] Cashmere Valley: A. W. *Uncinia rubra Boott. Abundant, 2,000 ft. and upwards, from Mount Sinclair to Castle Rock. Uncinia uncinata (Linn. f.) T. Kirk. [J. B. A.] Common in the forests and elsewhere. Uncinia leptostachya Raoul. [J. B. A.] Common in the forest, Akaroa (habitat of type). Uncinia riparia R. Br. var. Banksii C. B. Clarke. Mount Pleasant (det. Petrie). Akaroa: A. W. J. B. A. gives also (1)U. rupestris, which, being one of Raoul's species, is doubtless an inhabitant of the peninsula; but after a close search I am unable to find it. I think it unlikely that there are other forms on the peninsula than those listed above. If this be so, then U. rupestris of Raoul may be included in one of the forms above, possibly U. riparia var. Banksii; though it is to be admitted that none of my specimens coincide exactly with Raoul's illustration. Carex appressa R. Br. Near Lyttelton lighthouse: A. Wall A new record. Carex virgata Sol. [J. B. A.] Cashmere Valley, below Marley's Road: R. M. L. Stony Bay: J. Andersen! Carex secta Boott. [J. F. A.; J. B. A.; L. C.] Not uncommon in swampy places and by the sides of streams—e.g., Teddington; Charteris Bay; Taylor's Mistake (on the shore); Wainui.

Carex breviculmis R. Br. [J. B. A.] In drier ground, often near the hilltops. Carex ternaria Forst. f. [J. F. A.; J. B. A.; L. C.] Common in damper places—indeed, the most abundant species of Carex on the peninsula. *(1)Carex Colensoi Boott. Specimens from Castle Rock were identified by Petrie as probably belonging to this species. *°Carex Raoulii Boott. [J. B. A.] Akaroa: Raoul. *Carex Wakatipu Petrie. (Con. Cheeseman.) Castle Rock, 2,500 ft. Carex lucida Boott. [J. B. A.; L. C.] Common in damper spots on the Cashmere Hills, in the neighbourhood of Governor's Bay and Teddington, and possib y elsewhere. Carex litorosa Bailey. Heathcote: B. D. Cross. Carex dissita Sol. [J. B. A.; L. C.] In pastures and forest; not very common. *°Carex trifida Cav. [J. B. A.] Akaroa: Raoul. *Carex pumila Thunb. [J. B. A.] Wainui: R. M. L. Akaroa: Raoul. *Carex flava Linn. var. cataractae B. Br. [J. B. A.] Top of Castle Rock. Carex pseudo-cyperus Linn. Watercourse, Port Levy. Carex Forsteri Wahl. By the side of bush-streams, Kaituna, Port Levy; Summit Track beyond Kennedy's Bush, Governor's Bay. Some of the specimens have rather the character of C. semi-Forsteri. I have paid too little attention to the genus Carex to be able to give a complete list, consequently additional species are to be expected here. J. B. A. gives in addition (2)C. pyrenaica, (1)C. teretiuscula, (1)C. stellulata, (1)C. testacea, (1)C. inversa, (1)C. vacillans. Some of these probably occur. Family Palmeae. *Rhopalostylis sapida Wendl. & Drude. [J. B. A.] Once not uncommon, now becoming rare — Holmes Bay (probably now extinct); Laverick's; Okain's; Le Bon's; Waikerikikeri; Damon's; Palm Gully, Akaroa. (The southernmost limit on this coast.) Family Lemnaceae. Lemna minor Linn. [J. F. A.; J. B. A.] In pools. Family Restiaceae. Leptocarpus simplex A. Rich. [J. B. A.] Heathcote, in salt marshes: B. D. Cross. Teddington: R. M. L.

Family Juncaceae. Juncus pallidus R. Br.(?). (The identification is not quite certain, as my specimens are somewhat immature.) Port Levy: R. M. L. Probably also on Lyttelton Hills, near Dry Bush: A. W. Juncus pauciflorus B. Br. [J. B. A.; L. C.] Abundant in the tussock pastures. Juncus vaginatus R. Br. [J. F. A.; J. B. A.] Akaroa: T. Kirk. Lyttelton Hills. Juncus polyanthemus Buchen. [J. F. A.; J. B. A.; L. C.] Common in damper pastures. Juncus maritimus Lam. var. australiensis Buchen. [J. B. A.] Common in salt marshes. (According to B. D. Cross, extends as far south as Timaru.) Juncus bufonius Linn. [J. B. A.] Abundant. *Juncus planifolius R. Br. Akaroa Raoul. Mount Herbert, near top: A. Wall. Kaituna, in the wayside ditches: R. M. L. Juncus caespiticius E. Mey. var. bracteatus Buchen. Cashmere Valley: A. Wall. *Juncus novae-zealandiae Hook. f. French Peak, and in a slender state on the top of Castle Rock. J. B. A. gives also (2) J. scheuchzerioides and (2) J. holoschoenus, neither of which is likely to occur. Luzula campestris DC. vars. [J. F. A.; J. B. A.; L. C.] Family Liliaceae. Rhipogonum scandens Forst. [J. F. A.; J. B. A.; L. C.] Common in the forest in many places. Cordyline australis Hook. f. [J. F. A.; J. B. A.; L. C.] Common on open hillsides. *Cordyline indivisa Steud. [J. B. A.] Apparently the southernmost limit on this coast. Above 1,800 ft. Purau Line, Akaroa Summit Road, not common. (2)C. Veitchii Hort., given as an additional species by J. B. A., is merely a synonym of C. australis. Astelia nervosa Banks & Sol. [J. B. A.; L. C.] Abundant, chiefly in forest. (2)A. Solandri and (2)A. linearis are recorded by J. B. A., but have not been observed by any one else. Phormium tenax Forst. [J. F. A.; J. B. A.; L. C.] Common in gullies and at moister spots. *Phormium Cookianum Le Jolis. [J. F. A.; J. B. A.] On cliffs, and often in the open above the bush line. Possibly occurs on the Lyttelton Hills, though not definitely recorded from there. (1)Bulbinella Hookeri is recorded by J. F. A. and J. B. A., and perhaps occurs.

Arthropodium candidum Raoul. [R.; J. F. A.; J. B. A.; L. C.] Not infrequent in shaded gullies and in bush. Aylmer's Valley, track to lighthouse: A. Wall. Cashmere Hills; Castle Rock; &c.: R. M. L. Akaroa (habitat of type). J. B. A. records also (2)Dianella intermedia. I do not think it occurs. *°Iphigenia novae-zelandiae Baker. [J. F. A.] Banks Peninsula: Cockayne, in Cheeseman's Manual. Family Amaryllidaceae. Hypoxis pusilla Hook. f. [L. C.] Victoria Park, Cashmere Hills, and elsewhere. Difficult to observe unless in flower. Flowers April to June, and again in early spring. Family Iridaceae. Libertia grandiflora Sweet(?). [J. F. A.; J. B. A.] Abundant. The plant is larger in every way than the common forms of L. ixioides that I am acquainted with elsewhere, and may be true L. grandiflora; but though the capsules are often ½ in. in length I have measured none exceeding that dimension; the leaves, however, are often over 1/3 in. wide. There is, at any rate, only one species to be found here. The bracts are always long narrow-linear or linear-lanceolate and acuminate. Hooker (Handbook, p. 274) evidently regards this form as L. grandiflora, so I have retained the name. J. F. A., J. B. A., and L. C. give L. ixioides. J. B A. includes also (2)L. restioides Klatt and (2)L. micrantha A. Cunn. Family Orchidaceae. Earina mucronata Lindl. [J. F. A.; J. B. A.; L. C.] I am not certain that I have seen this species. It is certainly less common than the following. Earina suaveolens Lindl. [J. B. A.] Not uncommon in rocky clefts. J. B. A records also (1)Dendrobium Cunninghamii, but I have not seen it. *Spiranthes australis Lindl. Banks Peninsula: J. B. A.! Several specimens labelled as from Banks Peninsula are in the Canterbury Museum. In the Lyttelton Times of the 6th April, 1918, in the column “From Nature's Book”† Edited by J. Drummond. appears the following: “Mr. J. B. Armstrong … reports that Spiranthes australis was very common on the Lake Ellesmere flats before they were drained. Both the red and white variety may be found there still” Thelymitra longifolia Forst. [J. F. A.; L. C.] Castle Rock; Sugarloaf, &c.; but becoming less common. J. B. A. includes (2)T. uniflora, which I have not seen.

Microtis unifolia Reichenb. [J. B. A.; L. C.] Not uncommon. Prasophyllum Colensoi Hook. f. [J. F. A.] Castle Rock, above 2,500 ft.; Lyttelton(?): R. M. L. Pterostylis Banksii R. Br. Not uncommon in the forest and heath. *Pterostylis australis Hook. f. Waikerikikeri; Checkley's Bush, Akaroa; and probably elsewhere. Pterostylis graminea Hook. f. Waikerikikeri; Castle Rock; Lyttelton. (1)Pterostylis foliata Hook. f.(?). Heathcote Valley. One specimen only seen, collected by Miss E. Holdsworth, and thus identified by Cheeseman: “I believe, however, that it is P. foliata, of which it has the sheathing bracts and short lateral sepals.” (2)Cyrtostylis oblonga is recorded by J. B. A. *Caladenia minor Hook. f. [J. B. A.] Akaroa: Miss M. Fyfe! Corysanthes triloba Hook. f. [J. F. A.; J. B. A.; L. C.] Not uncommon in damp shady spots. *Corysanthes macrantha Hook. f. [J. F. A.; J. B. A.] Castle Rock. Gastrodia Cunninghamii Hook. f. [J. B. A.] Okute Valley, and probably elsewhere. It has been reported to me from Akaroa, and near Tai Tapu, but I have not seen specimens. Other orchids reported by J. B. A. are (2)Dendrbium pygmaeum, (1)Corysanthes rotundifolia, (2)C. rivularis, (2)Lyperanthus antarcticus, (2)Caladenia Lyallii; and by J. F. A., (2)Corysanthes oblonga. Class DICOTYLEDONEAE. Family Piperaceae. Macropiper excelsum Miq. [J. F. A.; J. B. A.; L. C.] Not uncommon in the lower forest, particularly near the sea. Here finds its southernmost limit. Family Fagaceae. *Nothofagus Solanderi Oerst. Summit of Long Bay Ridge, Akaroa, and extending downward for some distance to the eastward. Forms resembling N. Blairii are to be found, but require further examination. *Nothofagus cliffortioides Oerst. Forms apparently belonging to this species occur intermingled with N. Solanderi on the Long Bay Ridge. J. B. A. records it with a (?). *Nothofagus fusca Oerst. [J. B. A.] Ridges from Stony Bay to Damon's Bay. Some magnificent specimens are found here in the forest.

Family Urticaceae. Paratrophis microphylla Cockayne. [J. F. A.; J. B. A.; L. C.] Abundant in the forest. Urtica incisa Poir. [J. F. A.; J. B. A.; L. C.] In forest. Urtica ferox Forst. f. [J. F. A.; J. B. A.; L. C.] Abundant in forests and in their neighbourhood. *Parietaria debilis Forst. f. [J. B. A.] In scrub near Caton's Bay, Little Akaloa, and elsewhere, but not common. Australina pusilla Gaud. [J. B. A.] Mount Pleasant; Cooper's Knobs (perhaps extinct in this locality); Port Levy, on sides of small waterfall; Barry's Bay, in dark creek. Family Loranthaceae. Loranthus micranthus Hook. f. [J. F. A.; J. B. A.; L. C.] Common; becoming a pest in Akaroa, where it attacks fruit-trees and other deciduous trees such as Robinia pseudacacia, and even the willow. (See also Out in the Open, p. 135.) Tupeia antarctica Cham. & Schl. [J. F. A.; J. B. A.; T. P.; L. C.] Lyttelton (Out in the Open, p. 138); Stony Bay; Akaroa; Caton's Bay; Port Levy. Korthalsella Lindsayi Engler. [J. F. A.; J. B. A.; T. P.] On Myrsine Urvillei at Purau, on Myrtus obcordata at Stony Bay, and on Melicope simplex at Port Levy. Korthalsella salicornioides Van Tiegh. [J. B. A.; L. C.; T. P.] Wainui; Caton's Bay, on Leptospermum: R. M. L. Lyttelton Hills: Potts; Cockayne. Dover Castle: A. W. Of this species Potts (Out in the Open, p. 139) writes: “The writer only knows one habitat, that amongst a group of rocks just above the sea in Port Cooper. At the spot mentioned it makes use of the small-leaved manuka (Leptospermum ericoides) as a fostering plant.” Doubtless extinct in locality referred to. Family Polygonaceae. Polygonum aviculare Linn. [J. F. A.] By roadsides, common; probably introduced. Rumex flexuosus Soland. [J. F. A.; J. B. A.; L. C.] Common. Muehlenbeckia australis Meissn. [J. F. A.; J. B. A.; L. C.] Abundant. Muehlenbeckia complexa Meissn. [J. F. A.; J. B. A.; L. C.] Everywhere abundant at the edge of forests and in dry stony ground. *Muehlenbeckia axillaris. [J. B. A.] Top of Saddle Peak. Seen nowhere else by me.

Family Chenopodiaceae. Rhagodia nutans R. Br. [L. C.] Lyttelton Hills, amongst stones; Governor's Bay; Pigeon Bay, by seaside: R. M. L. Sleepy Cove, between Long Bay and Stony Bay, and doubtless elsewhere by the sea-coast: A. Wall. Chenopodium triandrum Forst. f. [J. F. A.; J. B. A.] Common in dry rocky ground. In specimens from the Lyttelton Hills the perianth-segments numbered five. Chenopodium glaucum Linn. [J. F. A.; J. B. A.] Taylor's Mistake. Chenopodium carinatum R. Br. Above Heathcote Reservoir. Probably introduced. Atriplex patula Linn. Heathcote: B: D, Cross. Ohinitahi; Teddington. J. F. A. and J. B. A. give also (1)Chenopodium ambiguum, which possibly occurs. Salicornia australis Linn. Abundant in salt marshes. Suaeda maritima Dum. The beach, Le Bon's Bay; Redcliffs: A. W. Family Aizoaceae. Tetragonia expansa Murr. Abundant on coastal banks, but occasionally ascending to 1,000 ft. on cliffs. Tetragonia trigyna Banks & Sol. Less common than the preceding. Also ascending to 1,000 ft., but usually on the sea-coast. Mesembryanthemum australe Sol. Common on the sea-coast. Family Caryophyllaceae. *°Gypsophila tubulosa Boiss. Lake Forsyth: Kirk. Stellaria parviflora Banks & Sol. [J. B. A.; L. C.] Common. A much depauperated form with only 3 or 4 stamens occurs at the altitude of about 1,000 ft. on the southern face of Mount Pleasant. Colobanthus Muelleri Kirk. Common in similar places to the following species. Colobanthus Billardieri Fenzl. [J. B. A.; L. C.] Not uncommon in rocky places near the hilltops. (1)C. quitensis Bartl. is also recorded by J. B. A., and may occur. Spergularia media Presl. Not uncommon on rocky coasts. Probably = S. marina var. rubra of J. B. A. Scleranthus biflorus Hook. f. [J. B. A.; L. C.] In dry and rocky ground; common.

Family Ranunculaceae. Clematis indivisa Willd. [J. B. A.; L. C.] At one time common in the forest; now much less so. C. hexasepala Forst. is recorded by J. B. A. and T. K., but does not occur. A specimen labelled C. hexasepala and collected by T. Kirk at Governor's Bay is in the collection at the Canterbury Museum, but, as pointed out on the sheet by L. C., is merely a well-grown plant of C. foetida. *Clematis Colensoi Hook. f. [J. B. A.] Price's Bush, near Birdling's Flat: S. Page! W. Martin. Clematis foetida Raoul. [R.; J. F. A.; J. B. A.; L. C.] Everywhere common in the forest, and in Birdling's Bush forming stems which rival those of C. indivisa in diameter. (Akaroa is the type-locality.) (2)C. parviflora is also recorded by J. B. A., but I have not seen it, and doubt its occurrence. Clematis afoliata Buch. [J. B. A.; L. C.] Not uncommon on the sea-coast, and occasionally on the drier cliffs up to 1,500 ft. At Sumner it is becoming more abundant. C. marata is not uncommon on the sand-dunes at New Brighton, but is probably not to be found on the peninsula. Ranunculus hirtus Banks & Sol. [J. B. A.; L. C.] Common in the tussock country. Ranunculus lappaceus Smith var. multiscapus Hook. f. [J. F. A.; J. B. A.; L. C.] Common in open country. Ranunculus rivularis Banks & Sol. var. subfluitans Benth. Pond, Cashmere Hills. J. F. A. and J. B. A. record also (2)R. pinguis, an Auckland Island plant, and J. F. A. gives (2)R. macropus. Family Magnoliaceae. Drimys colorata Raoul. [R.; J. B. A.; L. C.] Abundant, particularly on the upper edges of the forests. (Akaroa is the type-locality.) J. F. A. gives (2)D. axillaris, no doubt in place of the above species, and T. Kirk (Students' Flora) states the D. axillaris goes as far south as Banks Peninsula, but he is, I think, in error. Family Monimiaceae. Hedycarya arborea Forst. Common in the forest. Its southernmost limit on this coast, though it extends as far as Preservation Inlet on the west coast. Family Cruciferae. Cardamine heterophylla Schulz. [J. F. A.; J. B. A.; L. C.] Usually weak and decumbent; but a stout, erect variety (perhaps var. macrantha) occurs on the cliffs on the northern side of Mount Pleasant, above the Summit Track (not seen in flower). The weak and decumbent form may be var. leiocarpa. *Lepidium oleraceum Forst. f. var. acutidentatum Kirk. Coastal cliffs, Waikerikikeri.

Family Droseraceae. J F. A. and J. B. A. record, but probably in error, (2)Drosera spathulata and (2)D. binata. Family Crassulaceae. Crassula Sieberiana Schultz. [L. C.] An annual, abundant in dry, hard ground and on rocks in spring and summer. *Crassula moschata DC. [J. B. A.] Crown Island Run, on the coast. J. B. A. records also (2)Tillaea Sinclairii. Family Saxifragaceae. Carpodetus serratus Forst. [J. F. A.; J. B. A.; L. C.] A common forest tree. J. B. A. lists also (2)Quintinia serrata, a quite unlikely species. Family Pittosporaceae. Pittosporum tenuifolium Banks & Sol. [R.; J. F. A.; J. B. A.; L. C.] Abundant in forest, and here erroneously called “matipo.” The relative proportions of the leaves vary considerably in breadth and length in different specimens. J. B. A. and J. F. A. record also P. Colensoi, but a specimen so named by J. B. A. in the Canterbury Museum is a common form of P. tenuifolium. Pittosporum eugenioides A. Cunn. [J. F. A.; J. B. A.; L. C.] Common in the forest. *°(1)Pittosporum obcordatum Raoul. [R.; J. B. A.] Akaroa: R. (type-locality). Does not appear to have been found since the days of Raoul at Akaroa, though it appears in J. B. A.'s list (see Cheeseman, Trans., vol. 39, p. 436). Family Rosaceae. Rubus australis Forst. f. var. glaber Hook. f. [J. F. A.; J. B. A.; L. C.] Abundant in the forest. Rubus cissoides A. Cunn. [J. B. A.] In the forests; abundant. Var. pauperatus Kirk, though sometimes found in the forest, is more often found in the open, chiefly on the sites of old forests. [J. B. A.; L. C.] Rubus schmidelioides A. Cunn. [J. B. A.] var. coloratus Kirk [L. C.] Common on the margin of the forests, and in rocky ground above the forests. Rubus subpauperatus Cockayne. [L. C.] Common near the outskirts of the forest. Potentilla anserina Linn. var. anserinoides Raoul. [R.; J. F. A.; J. B. A. L. C.] In swampy ground occasionally. (Akaroa is the type-locality.)

°Geum urbanum Linn. var. strictum Hook. f. [J. F. A.; J. B. A.; L. C.] Tussock pastures: Cockayne. Acaena novae-zelandiae Kirk var. viridissima Bitter. [J. B. A.; L. C.] This is the most abundant form of Acaena on the peninsula, and is everywhere common up to 1,600 ft., particularly in dry open ground. In the forest it is usually replaced by the following, which, however, is by no means so common. Acaena Sanguisorbae Vahl. var. This variety is thus described from my specimens by L. Cockayne in a letter to me: “Plainly to be distinguished by the dark dull green and but slightly hairy upper surface of the leaf, the silvery undersurface with closely appressed hairs, the sepals green edged with purple within and still more hairy and purple without, the frequently trifid stipules, and the rather short stout spines, which vary from pale to rather dark purple.” This variety, comparatively rare on the Lyttelton Hills, occurs frequently on the peninsula above 1,500 ft., and at all levels in the forest. Acaena Sanguisorbae Vahl. var. pilosa Kirk. This form as described by Kirk occurs on the southern slopes of Castle Rock and Mount Herbert above the height of 2,000 ft., probably also at Cooper's Knobs. It is at once distinguished from the preceding by its glaucous bluish coloration, with brown serrations. The undersurface of the leaf, the upper margin, the petiole, peduncles, and stem are all markedly pilose. (I am indebted to Dr. Cockayne, who is making a special study of the New Zealand species of the genus, for this identification). A fourth form of Acaena (Acaena Sanguisorbae Vahl. var. viridior Cockayne) occurs in a small piece of bush in the Wainui Valley. J. B. A. gives also (2)A. adscendens, but I do not think it occurs. Family Leguminosae. *(1)Carmichaelia nana Col.(?). [J. B. A.] I have seen specimens belonging, I think, to this species near the top of Mount Herbert, on the northern side. Neither flowers nor fruit were present, and I was unable to identify it with certainty. C. nana is common on the old river-bed of the Waimakariri, seven miles from Christchurch. Carmichaelia subulata Kirk. [L. C.] Abundant in open country from the seashore to upwards of 1,500 ft. This is the only large species on the peninsula. J. B. A. records (2)C. australis, which does not come south of Pelorus Sound, (2C. flagelliformis (probably in error for the above), (2)C. pilosa Col.(?), and with J. F. A., in addition, (2)C. grandiflora. Sophora microphylla J. Müll. [J. F. A.; J. B. A.; L. C.] Common in the forest, and sometimes on the open hillside. Sophora prostrata J. Mull. [L. C.] Common in rocky ground, and here a very distinct species.

*Sophora tetraptera J. Müll var. A third and very distinct form of this genus occurs between Raupo and Stony Bay, and apparently has no juvenile stage. On submitting a specimen to T. F. Cheeseman he commented on it as follows: “This seems to be nearer to S. grandiflora than to S. microphylla, but does not match the East Cape plant, which must be taken as the type.” The leaves are about 10 cm. long with 15–20 pairs of linear-oblong leaflets, the standard about 5/6 the length of the wings, and not reflexed. J. B. A. records (2)S. grandiflora. Family Geraniaceae. Geranium microphyllum Hook f. [J. F. A.; J. B. A.; L. C.] Common in grasslands. Geranium sessiliflorum Cav. var. glabrum Knuth. [J. B. A.] Peak between Cass Peak and Cooper's Knobs; Barry's Bay; Castle Rock (2,500 ft.). Geranium dissectum Linn. var. australe Benth. [J F. A.; J. B. A] Not common, and usually in cultivated places. Pelargonium inodorum Willd. [J. F. A.; L. C.] Cashmere Hills and elsewhere, chiefly in cultivated ground. Family Oxalidaceae. Oxalis corniculata Linn. [J. F. A.; J. B. A.; L. C.] Everywhere abundant in open country. Oxalis magellanica Forst. Old Purau Road, Mount Fitzgerald (2,500 ft.). Family Linaceae. Linum monogynum Forst. f. [J. F. A.; J. B. A.; L. C.] Everywhere common on the coastal clifis, and sometimes also found on rocky crests and summits. The introduced L. marginale is sporadic and adventive. It was at one time common near the foot of Dyer's Pass Road. Family Rutaceae. Melicope simplex A. Cunn. [J. B. A.; L. C.] Everywhere abundant in the forest. Family Euphorbiaceae. Euphorbia glauca Forst. f. [J. F. A.; J. B. A.] The beach, Le Bon's: R. M. L. The beach, Wainui: A. W.; R. M. L. Family Callitrichaceae. Callitriche verna Linn. Wainui; common in streams. Apparently a new record.

Family Coriariaceae. Coriaria sarmentosa Forst. f. [J. B. A.; L. C.] Common in the tree form in the forest, and in the dwarf form in the open. The small form is considered to be more deadly to stock. Family Corynocarpaceae. Corynocarpus laevigata Forst. [J. B. A.] At one time a few scattered specimens probably existed along the coast from Dampier's Bay, Lyttelton, to Long Lookout Point. It has been suggested by J. B. A. that it is an escape from cultivation; but there is nothing in its distribution at Long Lookout Point, the only place where it now occurs, to suggest this. It there extends to a distance of a mile and a half inland from the beach, and has been comparatively abundant over this area. It is said that a grove existing at Macintosh Bay was felled by the owner in order to discourage the Maoris from visiting the place. One plant in Aylmer's Valley, Akaroa, found by Miss Fyfe! (See introduction for further account of its distribution.) Family Icacinaceae. Pennantia corymbosa Forst. [J. F. A.; J. B. A.; L. C.] Common in the forest. Family Sapindaceae. Dodonaea viscosa Jacq. [J. B. A.; L. C.] Common in the coastal forest—e.g., Gollan's Bay; Rapaki; Lake Forsyth, Akaroa. Alectryon excelsum Gaertn. [J. F. A.; J. B. A.; L. C.] Gollan's Bay; Port Levy; Birdling's Bush, where it forms a considerable proportion of the forest; and elsewhere. Its southernmost limit on this coast. Family Rhamnaceae. Discaria toumatou Raoul. [R.; J. F. A.; J. B. A.; L. C.] Abundant in open country. (Akaroa is the type-locality.) Family Rlaeocarpaceae. Aristotelia racemosa Hook. f. [J. F. A.; J. B. A.; L. C.] Abundant in the forest. *Aristotelia fruticosa Hook. f. [J. B. A.] Forest on southern side of Mount Herbert (2,500 ft.); only several specimens seen. *Elaeocarpus Hookerianus Raoul. [R.; J. B. A.] Caton's Bay; Tikao Bay: S. Page. Aylmer's Valley; Waikerikikeri; Port Levy. In most places only a solitary specimen or two. (Akaroa is the type-locality.) *(2)Elaeocarpus dentatus Vahl. [J. F. A.; J. B. A.] I have not certainly seen this on Banks Peninsula. It occurs in Deans's Bush, near Christchurch; and a specimen of Elaeocarpus at Pigeon Bay probably belongs to this species, but I have not seen it in flower.

Family Malvaceae. Plagianthus divaricatus Forst. [J. F. A.; J. B. A.] Common in salt-water marshes, as at Teddington; or on stony beaches—e.g., Port Levy. (1)Plagianthus cymosus T. Kirk. A specimen apparently belonging to this species was found on Mount Pleasant, behind Lyttelton, by Cockayne and Petrie; but in spite of numerous searches I have been unable to rediscover the plant. Cockayne now thinks it may have been a flowering plant of the juvenile form of P. betulinus. Plagianthus betulinus A. Cunn. [J. F. A.; J. B. A.; L. C.] Abundant in the forest. Hoheria angustifolia Raoul. [R.; J. F. A.; J. B. A.; L. C.] Common in the opener forest and by roadsides. (Akaroa is the type-locality.) Family Guttiferae. Hypericum gramineum Forst. f. [J. F. A.; J. B. A.; L. C.] Tussock-grasslands. Hypericum japonicum Thunb. [J. F. A.; J. B. A.; L. C.] Tussock-grasslands. Family Violaceae. Viola filicaulis Hook. f. [J. B. A.] Hilltops chiefly. Viola Cunninghamii Hook. f. [J. F. A.; J. B. A.; L. C.] Tussock-grassland and swamps. Melicytus ramiflorus Forst. [J. F. A.; J. B. A.; L. C.] Everywhere common. Known on the peninsula as “cow-leaf” or “whity-wood.” The North Island Maori name, “mahoe,” is quite unknown here, as is also the Otago “ini-ini.” (2) Melicytus lanceolatus is recorded only by J. B. A. I have not seen it nearer than Peel Forest. °(1)Melicytus micranthus Hook. f. var. microphyllus Cheesm. [J. B. A.; L. C.] Hymenanthera crassifolia Hook. f. [J. B. A.; L. C.] Abundant in rocky places, particularly near the hilltops. A rigid shrub with spinous divaricating interlacing branches, forming a mat closely appressed to stones and cliff-faces. The berries are found on the underside of the mat, and remain white when not exposed to light, but thereafter become dark blue—almost black. It has been regarded as a coastal plant, but is not so here. The species is extremely plastic, probably as much so as Discaria toumatou has been shown to be by Cockayne. In caves and moist forests it loses its thorny and divaricating character, forms long pliant branchlets pubescent towards the tips, and becomes clothed with somewhat larger leaves. In this form it is perhaps not different from H. dentata.

Family Passifloraceae. Tetrapathaea australis Raoul. [R.; J. F. A.; J. B. A.; L. C.] Becoming extinct on the Lyttelton Hills. Maori Pa, Port Levy; Pigeon Bay; Aylmer's Valley; Checkley's Bush, Akaroa, common; Barry's Bay; Caton's Bay. Here at its southernmost limit on this coast. Family Thymelaeaceae. *Drapetes Dieffenbachii Hook. Common above 2,000 ft. Leaves hairy at the tip but not on the margin. In specimens from Mount Brasenose the perianth-lobes are as long as the tube and externally villous. Dr. Berggren has identified this with the Tasmanian D. tasmanica, which it resembles,† Cheeseman, Manual, p. 165. but in the absence of specimens of the latter it must be left as it is. I have seen no pimeleas on the peninsula; but J. B. A. lists (2)P. prostrata, though J. F. A. gives it merely as an inhabitant of Canterbury swamps. Family Myrtaceae. Leptospermum scoparium Forst. [J. F. A.; J. B. A.; L. C.] Everywhere common in heaths. Leptospermum ericoides A. Rich. [J. F. A.; J. B. A.; L. C.] Abundant, and in the Kaituna Valley growing into trees more than 2 ft. in diameter at the bole and 50 ft. high. Metrosideros hypericifolia A. Cunn. [J. B. A.; L. C.] Not common on the Lyttelton Hills, but plentiful in the forest elsewhere on the peninsula. According to J. B. A., (2)M. scandens and (2)M. Colensoi also occur on the peninsula, but these records are almost certainly erroneous. Myrtus obcordata (Raoul) Hook. f. [R.; J. F. A.; J. B. A.; L. C.] Common in forest and scrub. (Akaroa is the type-locality.) *Myrtus pedunculata Hook. f. [J. B. A.] Kaituna Valley and elsewhere; but not nearly so common as the preceding. I do not think it occurs on the Lyttelton Hills. Family Onagraceae. *Epilobium Billardierianum Ser. [J. B. A.] “At sea-level in Sleepy Cove, between Long Bay and Stony Bay”: A. Wall. Epilobium junceum Sol. var. cinereum Haussk. [J. B. A.; L. C.] Lyttelton Hills, and elsewhere; common. Epilobium junceum Sol. var. cinereum Haussk. [J. B. A.; L. C.] Wainui, Cashmere Valley: R. M. L. Between Little River and Port Levy Saddle, in damp places: A. Wall. Epilobium pubens A. Rich. [J. B. A.; L. C.] In many places; common.

*Epilobium pictum Petrie. “On steep banks by the roadside in Balgueri Valley (Akaroa) and Pigeon Bay Peak”: A. Wall. *Epilobium alsinoides A. Cunn. Common on the summits from Mount Bossu to Mount Herbert; Waikerikikeri; Stony Bay Peak. * Epilobium tenuipes Hook. f. Mount Herbert, 2,000 ft.: A. Wall. *Epilobiwum insulare Haussk. On swampy hillsides from sea-level to nearly 3,000 ft., common—e.g., on the south side of Rocky Peak above the hilltop, and on the northern and eastern sides of Mount Herbert near the summit: A. Wall. Epilobium rotundifolium Forst. f. [J. F. A.; J. B. A.; L. C.] Common in moist places. Epilobium linnaeoides Hook. f. Not common. Heathcote Valley; Wainui: R. M. L. “In damp stony places. Bush at the head of Stony Bay; southern face of Mount Herbert, at about 2,500 ft.; Dan Rogers Gully”: A. Wall. Epilobium nummularifolium R. Cunn. [J. F. A.; L. C.] Abundant on damp rocks and banks. Epilobium nerterioides A. Cunn. [L. C.] Abundant in tussock-grasslands. *Epilobium macropus Hook. [J. B. A.] Southern side of Mount Herbert, near the top: A. Wall. Epilobium novae-zealandiae Haussk. Akaroa Summit Road, and elsewhere near the hilltops, and sometimes down to 500 ft. Very variable in form, and occasionally approaching E. glabellum. J. B. A. and J. F. A. record several other species which have been seen by no other observers on the peninsula—e.g., (2)E. purpuratum, (2)E. microphyllum, (2)E. crassum. Fuchsia excorticata Linn. [J. F. A.; J. B. A.; L. C.] Everywhere common in the forest. Fuchsia Colensoi Hook. f. [J. F. A.; J. B. A.; L. C.] Common in the forest and by the wayside. Family Halorrhagidaceae Halorrhagis erecta Schindler. [J. F. A.; J. B. A.; L. C.] Common in the open country and in scrub. *Halorrhagis depressa Walp. Common along the summits; everywhere above 1,500 ft. Myriophyllum intermedium DC. [J. F. A.] Cashmere Valley; Lake Ellesmere; Kaituna; Wainui. *Gunnera monoica Raoul. [R.; J. B. A.] On the ridge between Akaroa and Flea Bay; Saddle Hill; near summit of Mount Herbert; and in many other localities facing south and south-west, usually above 1,500 ft., but coming down in Wainui to less than 500 ft. (Akaroa is the type-locality.) The Gunnera monoica recorded by J. F. A. from the sandhills is doubtless (2)G. arenaria.

Family Araliaceae. Nothopanax arboreum Seem. [J. B. A.; L. C.] One of the most abundant of forest-trees. Sometimes called “figwood” by the settlers, or “New Zealand fig.” The surveyors' ugly name of the North is here unknown. *Nothopanax Colensoi Seem. [J. B. A.] Long Bay, in the beech forest, not common; Mount Herbert, in the subalpine scrub. Nothopanax anomalum Seem. Mount Pleasant (a few specimens only), and Caton's Bay. J. F. A. and J. B. A. record (2)P. simplex, but the record does not seem to have been confirmed. Schefflera digitata Forst. [J. F. A.; J. B. A.; L. C.] Abundant; called “ohau” at Akaroa. Pseudopanax crassifolium C. Koch var. unifoliolatum Kirk. [J. F. A.; J. B. A.; L. C.] Not uncommon in the forest. Pseudopanax ferox T. Kirk. [J. B. A.] Little Akaloa; Caton's Bay: R. M. L. Lake Forsyth: T. Kirk. Stony Bay: J. Andersen! And elsewhere, but quite uncommon. Family Umbelliferae. Hydrocotyle elongata A. Cunn. [J. B. A.; L. C.] Mount Pleasant; Barry's Bay; not common. Hydrocotyle americana Linn. [J. B. A.; L. C.] Lyttelton Hills; Stony Bay; and doubtless elsewhere. Hydrocotyle americana Linn. var. heteromeria Kirk. [J. B. A.; L. C.] Cashmere Valley. Hydrocotyle novae-zealandiae DC. [L. C.; J. B. A.] Lyttelton Hills. I have not seen this on Banks Peninsula, but doubtless it occurs. Hydrocotyle moschata Forst. f. [J. B. A.; L. C.] Lyttelton Hills (and doubtless elsewhere), common. The most abundant species of Hydrocotyle in the neighbourhood. Hydrocotyle microphylla A. Cunn.(?). As Cheeseman points out, this may not be Cunningham's plant; but a small and very distinct form of Hydrocotyle agreeing with the description of this species in the Manual is found very commonly on the tracks and in the bush, usually above 1,000 ft. I append a brief description: Leaves four- or five-lobed, to the middle or slightly beyond, about 5–7 mm. across, glabrous or occasionally slightly farinose, lobes sometimes again notched. Petioles rather long (15–20 mm.), sometimes with a few isolated hairs. Umbels with 2–5 rays, almost sessile, enclosed at the base by several translucent minute rotundate or broadly oblong bracts. Flowers almost sessile, but peduncle elongating in fruit to length of carpel. Fruit laterally compressed, with a rather prominent rib on each face. Carpels rounded at the back. This differs from H. americana in the rounded carpels and uniformly smaller size, and

from H. novae-zealandiae in the much smaller size and more deeply indented leaves. The species is very distinct in appearance and has a well-defined habitat. J. B. A. records also (1H. asiatica, (2)H. muscosa, and (2)H. dissecta. H. asiatica is certainly to be expected; but, though I have seen it growing freely as an imported weed in a garden on Cashmere Hills, I have nowhere seen it in a natural habitat. Schizeilema Hookeri Domin. [L. C.] Rather rare; usually at an altitude of 1,200 ft. or upwards, on southerly slopes or in forest, but in Checkley's Bush (Akaroa) and Wainui reaching to within several hundred feet of the sea. Also occurs at Mount Pleasant; Castle Rock; near Kennedy's Bush track; and elsewhere. J. B. A. gives also (2)Pozoa (i.e., Azorella) hydrocotyloides—a very unlikely plant to be found here. Apium prostratum Labill. [J. B. A.] Common on coastal rocks. Apium prostratum Labill. var. filiforme Hook. Purau, Teddington, &c., in salt marshes. *Oreomyrrhis andicola Endl. var. Colensoi Kirk. In small quantities on the top of Mount Herbert: A Wall! Crantzia lineata Nutt. [J. B. A.] In salt marshes, Heathcote Estuary, Kaituna, &c Aciphylla Colensoi Hook. f. Purau line, above 2,000 ft. Aciphylla squarrosa Forst. f. [J. F. A.; J. B. A.; L. C.] Once common; now rapidly disappearing, particularly in sheep-country. Anisotome aromatica Hook. f. [J. B. A.; L. C.] Banks Peninsula, common above 1,500 ft.; but very rare on Lyttelton Hills, Cooper's Knobs, The Tors. Anisotome Enysii (Kirk) Laing(?)† Report on Scenery-preservation, 1914–15, p. 14.. Dr. Cockayne considers this may not be the same as the Castle Hill plant. Usually above 1,500 ft. Dover Castle; Mount Sinclair; Mount Berard; Purple Peak; One Tree Hill. J. B. A. records (2)Ligusticum piliferum, probably in error. Angelica montana Cockayne. [J. B. A.; L. C.] Once common; now only in rocky clefts and on shelves inaccessible to sheep. Angelica geniculata Hook. f. [J. B. A.; L. C.] Common, usually near the sea-coast. Lyttelton; Akaroa; Kaituna; Waikerikikeri; &c. °Angelica rosaefolia Hook. Akaroa: recorded by Raoul, but probably not seen since, though it appears in J. B. A.'s list. Daucus brachiatus Sieb. [J. B. A.; L. C.] Lyttelton Hills; Castle Rock.

Family Cornaceae. Corokia Cotoneaster Raoul. [R.; J. B. A.; L. C.] Abundant in rocky places near the hilltops, and more rarely on the coast, as at Island Bay. (Akaroa is the type-locality.) Griselinia littoralis Raoul. [R.; J. F. A.; J. B. A.; L. C.] Abundant everywhere in the forest. (Akaroa is the type-locality.) *Griselinia lucida Forst. f. Waikerikikeri (rare); Little River (perhaps extinct); Stony Bay; sometimes on black or white pines. Probably its southernmost limit. Family Ericaceae. Gaultheria antipoda Forst f. [J. B. A.] Banks Peninsula, 2,000 ft. and upwards; usually prostrate and creeping, and small-leaved. Identified (probably wrongly) with var. depressa (Trans., vol. 46, p. 59). Lyttelton Hills (the erect form), on peak between Cass Peak and Cooper's Knobs; and at a low level in Wainui, in manuka scrub. J. B. A. gives also (2)G. rupestris, which I have not seen. Family Epacridaceae. *Pentachondra pumila R. Br. Near the top of Brasenose (perhaps extinct): R. M. L. Stony Bay Peak: A. Wall. Cyathodes acerosa Sol. [J. B. A.; L. C.] Not uncommon amongst rocks. Leucopogon Fraseri F. Muell. [R.; J. B. A.; L. C.] 1,500 ft. and upward, not common; but coming down to sea-level at Redcliffs. Leucopogon fasciculatum A. Rich.(?). [J. B. A.] Rapaki; Kaituna; Akaroa; and elsewhere. Dracophyllum acicularifolium Cockayne. [J. B. A.] Abundant on rocky ledges, 1,600 ft. and upwards; Cooper's Knobs, and peak to eastward. Family Myrsinaceae. Myrsine Urvillei A. DC. [J. F. A.; J. B. A.; L. C.] Everywhere common in the forest. Myrsine divaricata A. Cunn. [J. B. A.] Kennedy's Bush (one or two specimens); Le Bon's; Mount Fitzgerald; Kaituna; &c.; usually 1,200 ft. and upward. J. B. A. gives also (2)M. nummularia; but I doubt its occurrence. Family Primulaceae. Samolus repens Pers. var. procumbens R. Knuth. [J. F. A.] Coastal cliffs and beaches; common.

Family Gentianaceae. °Sebaea ovata R. Br. [J. F. A.; J. B. A.] Port Cooper: Lyall. Lake Ellesmere: T. Kirk. *Liparophyllum Gunnii Hook. f. Lake Ellesmere. J. B. A. includes (2)Gentiana montana Forst., but almost certainly in error Family Apocynaceae. Parsonsia heterophylla A. Cunn. [R.; J. F. A.; J. B. A.; L. C.] Abundant in the forest. Parsonsia capsularis R. Br. var. rosea Cockayne. [R.; J. F. A.; J. B. A.; L. C.] Very common in the forest; often more abundant than the preceding. (Akaroa is the type-locality of the variety.) Family Convolvulaceae. Calystegia tuguriorum R. Br. [J. F. A.; J. B. A.; L. C.] Not uncommon in the forest. C. sepium occurs in abundance by the waysides and in gardens, but is probably introduced. Calystegia Soldanella R. Br. [J. B. A.] On sandy beaches. Convolvulus erubescens Sims. Common on dry hillsides, particularly those exposed to the sea-wind. Dichondra repens Forst. [J. F. A.; J. B. A.; L. C.] Less common than the following. Dichondra brevifolia Buch. Abundant in dry open ground, and often found in gardens (Cashmere Hills). The flowers at the tips of the branches are almost sessile. Family Boraginaceae. (1)Myosotis spathulata Forst. f. [J. B. A.] A form from Mount Pleasant has been so identified by Petrie; but further specimens are required. Myosotis pygmea Col. [J. B. A., as M. antarctica(?).] Gollan's Bay; Sugarloaf; behind Governor's Bay; usually above 1,000 ft. Myosotis australis R. Br. var. [J. B. A.; L. C.] Lyttelton Hills, southern side, at the foot of cliffs and on rocky ledges, usually about 600–800 ft. This plant has been thus identified by T. F. C. and L. C.; while Petrie has hesitated to identify it with any known species. It is undoubtedly intermediate between M. australis and M. Forsteri, and where it not for the weight of authority against me I should consider it more closely allied to M. Forsteri than to M. australis. It differs from M. australis in the almost prostrate stems, less hispid stem and leaves, shorter racemes; the newly ripe seeds are polished and shining, dark brown, not black. It differs from M. Forsteri in being much stouter, the pedicels are shorter and do not

equal the calyx, the leaves are spathulate rather than orbicular-oblong or oblong, being one and a half to twice as long as broad. The nutlets are ovoid, not orbicular. The colour when dry is not so dark as that of M. australis, nor so green as that of M. Forsteri. The flowers are invariably white, the calyx tubular in the flower, becoming campanulate in the fruit. I propose, however, to leave the discussion of this and other critical species for a subsequent paper. *Myosotis Forsteri Lehm. [J. B. A.] A more typical form of M. Forsteri occurs on the Akaroa — Flea Bay Ridge above 2,000 ft. J. B. A. has also (2)M. capitata. Family Verbenaceae. Teucridium parvifolium Hook. f. [J. F. A.; J. B. A.; L. C.] Mount Pleasant; Caton's Valley. Perhaps now extinct in the former locality. Family Labiatae. Mentha Cunninghamii Benth. [R.; J. F. A.; J. B. A.] Hoon Hay Valley; hills behind Tai Tapu. (2)Scutellaria has been recorded by J. A. B., but probably in error. Family Solanaceae. Solanum aviculare Forst. f. [J. F. A.; J. B. A.; L. C.] A common weed in country that has been recently cleared of bush, and in the outskirts of the forest; not common elsewhere. Solanum nigrum Linn. [J. F. A.; L. C.] An abundant weed in gardens and elsewhere, appearing as an introduction. Family Scrophulariaceae. Mimulus repens R. Br. Heathcote Valley; Kaituna (Lake Ellesmere). °(1)Mazus pumilio R. Br. [Lyall; J. B. A.] A doubtful inhabitant. (2)Gratiola nana is also recorded by J. B. A. *Limosella tenuifolia Nutt. Lake Ellesmere. Specimens collected by T. Kirk under the name L. aquatica are in the Canterbury Museum. Veronica salicifolia Forst. f. var. communis Cockayne. [J. B. A.; L. C.] Abundant in scrub and near the edge of the forest. Veronica leiophylla Cheesem. [L. C.] This is the V. ligustrifolia Cunn. of Armstrong's list. It is abundant on the upper fringes of the bush, and is often not found elsewhere. In the Kaituna Valley it occurs all through the forest, reaching from over 2,000 ft. down to sea-level; in other places where it comes down to sea-level (e.g. Akaroa, Crown Island, Little Akaloa) the leaves become shorter and broader than in the typical form.

*Veronica buxifolia Benth (forma). One plant only seen, on Summit Track in Greenland Bush, Mount Herbert. (2)V. pimeleoides, reported by Lyall from Port Cooper, may possibly, though not probably, be V. Lavaudiana. At any rate, nothing resembling V. pimeleoides occurs near Lyttelton. Veronica Lavaudiana Raoul. [R.; J. F. A.; J. B. A.; L. C.] Everywhere abundant on rocky faces above 800 ft. or 900 ft. I have never seen it on the plains. Travers's record of it from there is probably an error. (Akaroa is the type-locality) *°Veronica Raoulii Hook. f. [R.; J. B. A.] I cannot help thinking that there must be some error in the record of a Banks Peninsula habitat for this species. It is said by Hooker to have been received by him from Raoul with specimens of V. Lavaudiana; but it seems improbable that such a careful botanist as Raoul should have confused two such different species. J. B. A., no doubt following Hooker, records it as from the peninsula. It is not, however, a plant which is likely to have become extinct. Growing in similar situations to V. Lavaudiana, one might have expected it to have been found in conjunction with it; but I have hunted assiduously for it in all likely places and have been unable to find it. The nearest point to the peninsula where I have found it is at White Rock, behind Rangiora, where a reduced form occurs. *Veronica Lyallii Hook. f. (forma).° Mount Fitzgerald, on the southern face of the big cliff near the summit. *°Veronica canescens Kirk. Lake Forsyth: Kirk. (2)V. cupressoides, (2)V. lanceolata, (2)V. vernicosa, and (2)V. stricta (a variety of V. salicifolia) are also recorded by J. B. A. V. cupressoides probably does not occur on the peninsula. V. lanceolata is a variety of the well-known V. catarractae, and is also quite unlikely to occur. Both J. F. A. and J. B. A. list (2)V. Colensoi, a very unlikely species. *Ourisia macrophylla Hook. f. (forma).† Trans., vol. 46, p. 59. [J. B. A.] Abundant, often covering the ground with matted patches, from about 2,300 ft. upwards. It occasionally extends into the upper edge of the forest, when it becomes more erect. *Euphrasia zelandica Wettst. Southern side and top of Mount Herbert and Castle Rock, 2,500 ft. and upward. Also found by A. W. J. B. A. records (2)E. Monroi, but it is quite unlikely that it occurs. Family Lentibulariaceae. *Utricularia monanthos Hook. f. [J. B. A.] Lake Ellesmere: J. B. A. There are specimens from this locality in the herbarium of the Canterbury Museum. At one time it occurred in a bog in the neighbourhood of Christchurch.

Family Myoporaceae. Myoporum laetum Banks & Sol. [J. F. A.; J. B. A.; L. C.] Abundant near the sea-coast, and sometimes inland. Family Plantaginaceae. Plantago Raoulii Decaisne. [R.; J. F. A.; J. B. A.] Seen by me, but exact locality overlooked: R. M. L. (Akaroa is the type-locality.) *Plantago spathulata Hook. f. Summit of Mount Herbert: A. Wall! Plantago Brownii Rapin. [J. B. A.] Heathcote Valley. Family Rubiaceae. The genus Coprosma has caused me much trouble, and I am not quite satisfied as yet that the list for the peninsula is complete. I have to thank Mr. Petrie for much kind assistance here. *Coprosma grandifolia Hook. f.(?). A few plants which perhaps belong to this species occur on the creek in R J. Fleming's place at Port Levy, and also on the sides of the main creek at Pigeon Bay. On submitting specimens to Dr. Cockayne he declared them to be typical C. grandifolia This determination, of course, gives a large southerly extension to the range of the species. A very similar form occurs in Caton's Bay, and also at Gore Bay, on the northern side of Pegasus Bay. The leaves are membranous, but little glossy, and range up to 8 in. long; and the peduncles are over 2 in. in length. In spite of this it appears to me probable that these are merely specimens of C. lucida of luxuriant growth. Growing in the warm, sheltered valleys of Port Levy and Pigeon Bay, under the shelter of the forest, the leaves have become more membranous, the whole plant has become more elongated, and the leaves are confined to the ends of the twigs. This seems to me to be more particularly so as intermediates may be found as in Caton's Valley and on the back of the Sugarloaf. However, the matter must be left for fuller future discussion. In spite of the fact that Cheeseman considers C. grandifolia one of the most distinct species of the genus, it seems to me that C. lucida grown under the same conditions would approximate closely to it or become identical with it. Coprosma lucida Forst. f. [J. F. A.; J. B. A.; L. C.] Common in the forests and on rocky promontories. Coprosma robusta Raoul. [R.; J. F. A.; J. B. A.; L. C.] Akaroa; Stony Bay; Lyttelton; Little River; &c. (Akaroa is the type-locality.) Coprosma Cunninghamii Hook. f. [J. F. A.; J. B. A.; L. C.] Common and very variable, extending from seashore to 1,500 ft. or 2,000 ft.; usually in scrub. Coprosma rotundifolia A. Cunn. [J. F. A.; J. B. A.; L. C.] Common in forest. Coprosma areolata Cheesem. [L. C.] Not uncommon—Lyttelton Hills; Port Levy; Pigeon Bay; and elsewhere. Varies considerably in the amount of pubescence on the leaves.

Coprosma rhamnoides A. Cunn. [J. B. A.; L. C.] Common in the forest and in scrub. *Coprosma parviflora Hook. f. [J. B. A.] Mount Sinclair; Brasenose; about 2,000 ft. and upwards. Coprosma crassifolia Col. [L. C.] Pigeon Bay; Akaroa; Mount Sinclair; Lyttelton; and generally common on open stony hillsides that have once been bushed. (1)Coprosma rigida Cheesem.(?). Mount Pleasant; Otahuna. Identification perhaps not quite certain Coprosma rubra Petrie. Near Cooper's Knobs: R. M. L. Lyttelton: A. W. Coprosma virescens Petrie. [T. K.] Pigeon Bay; Mount Pleasant; Lake Forsyth. With yellowish-green branchlets in the forest, but passing out into the open at Lake Forsyth, where it forms matted bushes that in autumn have brilliant red almost leafless twigs. *Coprosma acerosa A. Cunn. var. arenaria Kirk. [J. F. A.; J. B. A.] On the coast near Long Lookout Point. Coprosma propinqua A. Cunn. [J. B. A.; L. C.] Lyttelton Hills; Kaituna; and elsewhere; in many situations. Coprosma linariifolia Hook. f. [J. B. A.; L. C.] Common in the forest, particularly above 1,000 ft., but occasionally coming down to sea-level. J. F. A. gives also (2)C. spathulata, an Auckland species; (2)C. cuneata, usually an alpine plant; and (2)C. foetidissima, which might be expected, but does not, I think, occur. *Nertera depressa Banks & Sol. [J. B. A.] Castle Rock, boggy ground, 2,000 ft.: R. M. L. Western side of Mount Herbert, near the summit: A. W. *°(1)Nertera dichondraefolia Hook. f. Akaroa: Raoul (as N. gracilis Raoul). Galium umbrosum Sol. [J. B. A.] Common in stony ground, usually near the outskirts of the forest. Galium tenuicaule A. Cunn. [J. B. A.] Swamp, Cashmere Valley. Asperula perpusilla Hook. f [J. B. A.] Kennedy's Bush track; One Tree Hill. Family Campanulaceae. °Pratia angulata Hook. f. [J. B. A.] Lyttelton Hills: Cockayne Between Stony Bay and Flea Bay, damp places: A. W. *Lobelia anceps Linn. f. [J. F. A.; J. B. A.] Island Bay, seashore; coastal cliffs, Waikerikikeri: R. M. L. Children's Bay (Akaroa) and sea-coast generally: A. W. Wahlenbergia gracilis A. DC. [J. F. A.; J. B. A.; L. C.] Abundant in tussock-grasslands.

Wahlenbergia albomarginata Hook. (= W. saxicola of Cheeseman's Manual). Mount Sinclair; Castle Rock; above 2,500 ft.: R. M. L. Akaroa hilltops, common: A. W. A coastal form is said to exist under the cliffs at Lyttelton Heads, but I have not seen it. Family Goodeniaceae. Selliera radicans Cav. [J. F. A.; J. B. A.] Salt marshes, Teddington and elsewhere. Family Candolleaceae. Forstera tenella Hook. f. Abundant on the south and south-west faces at the head of Stony Bay and adjacent heights, western face of Mount Herbert: A. W. This species, found by Professor Wall, must be added to the list of the subalpine inhabitants of the peninsula, together with several other plants found by the same botanist: e.g., Epilobium tenuipes, E. macropus, Oreomyrrhis andicola. Family Compositae. *Lagenophora pumila Cheeseman. [J. F. A.; J. B. A.] Checkley's Bush; Akaroa; and probably elsewhere. Lagenophora petiolata Hook. f. [J. F. A.; J. B. A.; L. C.] Common in the tussock-grasslands. *Lagenophora pinnatifida Hook. f. var. hirsutissima Cockayne. Ridge between Waikerikikeri and Le Bon's, plentiful; a few plants on the top of French Peak and on the Long Bay Ridge. I have to thank Mr. Cheeseman for the determination. The rays are in more than one series; but I have not seen any ripe achenes. The plant agrees well with the description. Brachycome Sinclairii Hook. f. [J. F. A.] One of the forms of this very variable plant is abundant in the district It has been determined as B. Thomsoni by Cockayne,† Report on Scenery-preservation, 1914–15, p. 15. but all the specimens I have examined have radical leaves only. The plants on the Lyttelton Hills are usually small, and the leaves often but little lobulated, though membranous rather than fleshy. It seems to me to be distinct from B. Thomsoni as I have seen it in Stewart Island; and it is quite distinct from a subalpine form to be found in south Nelson with fleshy shining green entire leaves and large flowers. Olearia arborescens Cockayne and Laing. [J. B. A.] Some flowerless specimens have been seen on Mount Herbert by Professor Wall. It has also been recorded by J. F. A. from the Dry Bush, where it does not now exist. *Olearia ilicifolia Hook. f. Purau line, and occasionally elsewhere on the outer fringes of the bush above 2,000 ft. *Olearia ilicifolia Hook. f. In similar positions to the preceding, but even less common. Olearia avicenniaefolia Hook. f. [R.; J. B. A.] Akaroa: Raoul. Coastal scrub, Pigeon Bay: R. M. L. At one time it grew under the hill-crests behind Governor's Bay, but is perhaps now extinct there.

Olearia Forsteri Hook. f. [R.; J. F. A.; J. B. A.] Abundant, especially near the coast-line. Olearia fragrantissima Petrie. [L. C.] Decanter Bay; Mount Pleasant; Crown Island: R. M. L. Lake Forsyth: T. Kirk. Near Sumner-Lyttelton Road: A. W.; L. C. This plant does not appear to be known north of Banks Peninsula. *Olearia Hectori Hook. f.(?). I obtained specimens of a plant which may belong to this species near Le Bon's, but it was not in flower, and the identification is uncertain. J. B. A. records (1)O. virgata, and it may occur. It grows, or grew, on the banks of the adjacent Heathcote River, but I have not seen it on the hills. Celmisia longifolia Cass. var. gracilenta (Hook. f.) Kirk. [J. F. A.; J. B. A.; L. C.] The slender scape and narrow leaves (about 5 mm. broad) probably are sufficient marks of this variety. The upper surface of the leaf is dark, almost black, in colour; and often flat, not revolute, this character probably depending upon the amount of moisture received; and the flowers are often large, being fully 40 mm. in diameter. The lower half of the margin of the leaf is slightly toothed and repand, whilst the upper half is entire. I can scarcely think that the branching of the scapes referred to by Cockayne (Trans., vol. 49, p. 58) is at all common, as it has not been noticed by me. At one time common in the tussock lands, but now, I think, scarcer. Celmisia Mackaui Raoul. [R.; J. B. A.] In various places in the neighbourhood of Akaroa, from sea-level to 2,000 ft.; usually growing on a damp cliff or near the side of a stream. (Akaroa is the type-locality.) J. B. A. records also (2)C. coriacea, (2)C. Lyallii, and (2)C. spectabilis. There is no trace of them now, and I cannot think they have occurred here recently. Vittadinia australis A. Rich. [J. F. A.; J. B. A.; L. C.] Abundant. Vittadinia australis A. Rich. var. linearis Kirk, a very distinct form, also occurs between Lyttelton Heads and Heathcote Valley, but is probably an introduction. Gnaphalium luteo-album Linn. [J. F. A.; J. B. A.; L. C.] Abundant both as a garden-weed and as an inhabitant of the tussock-grasslands. Gnaphalium japonicum Thunb. [J. F. A.; J. B. A.; L. C.] Common. Gnaphalium collinum Labill. [J. B. A.; L. C.] Common. Raoulia glabra Hook. f. [L. C.] Lyttelton Hills, rare; behind Hoon Hay; and elsewhere abundant above 2,000 ft. *Raoulia subsericea Hook. f. On the higher summits between Akaroa and Castle Rock: R. M. L.; A. W.

Raoulia lutescens Kirk. One specimen only seen, on the new Lyttelton-Sumner Road, about 600 ft.; possibly from seed blown up from some river-bed on the plains. *Raoulia australis Hook. f. [R.; J. F. A.; J. B. A.] A rare and disappearing species; one specimen only seen, on the top of Purple Peak. (Akaroa is the type-locality.) Raoulia Monroi Hook. f. Dyer's Pass: R. M. L.; A. W. Mount Herbert and elsewhere: A. W. J. B. A. also records (2)R. tenuicaulis, but I have not seen it. Helichrysum bellidioides Willd. [J. F. A.; J. B. A.; L. C.] Lyttelton Hills, rare; but elsewhere not uncommon above 1,500 ft. Helichrysum filicaule Hook. f. [J. F. A.; J. B. A.; L. C.] Abundant. Helichrysum glomeratum Benth. & Hook. [R.; J. F. A.; J. B. A.; L. C.] *Cassinia Vauvilliersii Hook. f. [J. F. A.; J. B. A.] In scrub, Mount Herbert and elsewhere, but not common. J. B. A. gives also (1)C. fulvida, a probable inhabitant, which may be found on sandy beaches, but I have not noted it. J. F. A. records, no doubt in error, (2)C. leptophylla. Craspedia uniflora Forst. f. (one or more vars.). Lyttelton Hills, and probably elsewhere, but less common than formerly. J. F. A. and J. B A. give also (2)C. alpina, but this is a most unlikely inhabitant of the peninsula. Cotula coronopifolia Linn. [J. F. A.; J. B. A.] Common in ditches and sluggish streams. Cotula australis Hook. f. Abundant, and becoming a pest in gardens. Appears to be introduced. Cotula minor Hook. f. Seen only on Mount Pleasant, at the back of Lyttelton, under the drip from overhanging rock or on moist faces. Cotula Haastii Kirk. [Haast; T. K.; L. C.] Common on dry banks, particularly near the sea, or on the seaward side of the hills. It is said to have been found on the Canterbury Plains by von Haast, but I have not seen it there. The common species in the neighbourhood of Christchurch amongst the pastures are C. dioica and C. squalida. J. B. A.'s records of (2)C. pectinata as on the peninsula perhaps refers to C. Haastii. Cotula squalida Hook. f. [J. B. A.] Common in the tussock-grasslands. Cotula dioica Hook. f. [J. B. A.; L. C.] Two forms are common, one of the salt marshes and one of the hills; but it is probably useless to attempt at present to define them. J. B. A. gives also (2)C. pyrethrifolia, an unlikely alpine plant. Erechtites prenanthoides DC. [L. C.] Common, especially in burnt bush.

Erechtites quadridentata DC. [J. F. A.; J. B. A.; L. C.] Common by the wayside and in dry ground. *Erechtites glabrescens Kirk. Stony Bay; Kaituna; and doubtless elsewhere. *Erechtites arguta DC. [J. F. A.; J. B. A.] Little Akaloa. Senecio saxifragoides Hook. f. [J. B. A.] Abundant, Lyttelton Hills from Gebbie's Pass to Lyttelton North Head, particularly on the southerly faces of the hills from about 600 ft. and upwards; occasionally coming down to 300 ft. For a discussion of the characters and distribution of this species see article in Trans., vol. 50, p. 198, by Professor A. Wall Senecio lagopus Raoul. [R.; J. F. A.; J. B. A.] On or near the summits almost continuously from Akaroa Heads to Gebbie's Pass. In a few places near Gebbie's Pass it crosses on to the Lyttelton Hills. (Akaroa is the type-locality.) I am reported by Wall (loc. cit.) as stating that S. saxifragoides occurs on the peninsula. I now wish to withdraw that statement, as it was due to an imperfect understanding of the differences between the two species. There are several forms of S. lagopus to be found on the Canterbury Plains, which differ considerably from the type form at Akaroa, and the inclusion of these in my idea of the type led me into error. Undoubtedly the species lagopus as at present constituted contains several subspecies which require differentiation. J. F. A. records also (2)S. bellidioides. Typical bellidioides does not seem to occur on the peninsula, though young forms of S. lagopus are often indistinguishable from it. Senecio lautus Forst. f. [J. F. A.; J. B. A.; L. C.] Common in fairly constant forms, sea-level to 1,500 ft. Senecio sciadophilus Raoul. [R.; J. B. A.; L. C.] Sea-level to 2,000 ft.; not common. I have seen one plant at Kennedy's Bush, and it is fairly common on the Summit Track from behind Robinson's Bay to Mount Sinclair. It occurs more rarely in the lowland forest, as in Hay's Reserve, Pigeon Bay, Balgueri Valley, &c, It was reported from the Dry Bush (Lyttelton Hills), but is probably now extinct there. (Akaroa is the type-locality.) J. B. A. includes also (2)S. odoratus var. Banksii, a most unlikely plant. Microseris scapigera Sch. Bip. [J. F. A.; J. B. A.; L. C.] Lyttelton Hills: L. C. Castle Rock: R. M. L. Very uncommon here °Taraxacum magellanicum Comm. Lyttelton Hills: L. C. J. F. A. records (2)Crepis novae-zelandiae, which probably does not occur. Sonchus oleraceus Linn. Abundant, but doubtfully native. *Sonchus asper Hill var. littoralis Kirk. Sea-cliffs at Pigeon Bay; coast near Nikau Palm Gully; and doubtless elsewhere. Quite distinct from the British Sonchus asper. J. B. A. records (1)S. asper—the typical form, I presume—which, no doubt, occurs; but I have not as yet identified it here.