probably comes true from seed. For it I propose the name “canescens,” which would either be a variety of C. albida, if that be considered a species, or form canescens of C. vauvilliersii. Whatever view be taken matters little so long as so distinct a plant has a name. When growing amongst the scrub Cassinia albida is a shrub of slightly straggling habit, having many branches prostrate towards the base and then ascending, but quite naked except towards their apices. Here are given off from the main axis numerous close, short, erect shoots, ± 13 cm. in length, their lower halves unbranched and densely leafy, but above furnished with many short lateral shoots, given off from the stem at an acute angle, the length of the basal and largest shoots being ± 2·7 cm. Such shoots are also all densely leafy, and the whole erect dense head of foliage may be about 22 cm. long by 20 cm. broad. The leaves are closely imbricating in the bud, but when fully developed are patent, the lamina being bent at right angles to the short petiole. In size they measure about 1·1cm. to 1·4 cm. in length by about 4 mm. in breadth, which is larger than the measurements given by Kirk and Hooker for C. vauvilliersii The var. canescens, at least in my specimens, has rather longer leaves still—1·6 cm. The margins are sometimes faintly recurved. On the under surface of the leaf is a dense tomentum of matted, fine hairs. The shoot-axes are also densely tomentose. The young leaves are extremely glutinous, and give off a somewhat aromatic odour, which is quite evident as one passes through the scrub.

Cassina albida is not only found in the subalpine scrub, but descends to the river-flats on the Kaikoura Plain, where it would probably be more common but for the close growth of Leptospermum scoparium, with which in that situation it can ill compete. This is also shown on the mountain, where if a portion of the forest or scrub be burnt a Leptospermum heath takes possession, and becomes so strong a feature of the landscape that one might easily take it for a primitive formation. C. albida grows readily from seed, and in some parts of Mount Fyffe the montane meadow contains colonies of this plant. In such places are other forms of Cassinia—referrable perhaps to C. fulvida or to C. vauvilliersii—which form a bewildering mixture; or it may be that C. vauvilliersii is in a state of mutation, and that new forms are coming into being at the present time.

Olearia coriacea, of which only two adult and one juvenile specimen were noted, is a plant of considerable interest. During my former visit to Mount Fyffe I had not noticed it, but Kirk had found it there and also in the Awatere district. Notwithstanding that he had no flowering specimens, he described it in the

“Students' Flora,” p. 276, as he considered it to be a distinct species quite unlike any other. Olearia coriacea is a rather tall shrub, having few rigid rather tortuous branches. Below these branches are naked, but above they branch into short densely leafy twigs. Such leafy part is ±18 cm. long. The leaves are small, extremely coriaceous, shortly petiolate, and covered on their under surface with a very dense tomentum, white with a faint brownish tinge. The upper surface is yellowish-green, strongly marked by the reticulating veins, which give rise on the under surface to distinct lacunæ. The most striking feature about the leaf is its curious curving, the margins being so recurved and the leaf so arching upwards as to resemble a saddle, the apical portions being so recurved as to almost meet, while this part of the leaf curves upwards, ending in a point and thus forming the front of the saddle, the leaf-base being broad and rounded, and the centre forming the hollow. There were no signs of the young inflorescence, but the remains of those of last year showed that it is a short panicle about 3·5 cm. long. In shape, when flattened out, the leaves are ovate or broadly ovate, and the blade measures from about 2·1 cm. by 1·3 cm. to about 1·4 cm. by 8 cm. Growing near by was a plant of Olearia forsteri with small leaves, and one plant of a form of O. cymbifolia with large leaves, and this suggested both to Mr. Matthews and myself that perhaps the plant under discussion may be a hybrid between two latter species.

Phormium cookianum, which comes next in abundance to the dominat Cassinia, is a frequent important constitutent of subalpine scrubs. Its powers of resisting drought are very great, since it often grows also on faces of rock in close vicinity to the sea.

The two Coprosmas and Aristotelia fruticosa are common members of subalpine and river-terrace scrub, having the windshorn habit of so many New Zealand shrubs. The branches are interlacing, and the leaves small. The plants noted of the Aristotelia were not of the extreme xerophytic type that the species frequently assumes, showing in no case the reduction of leaves and the semi-spinous shoots that I have recorded else-where.^*

Dracophyllum uniflorum, like some others of its congeners, has erect needle-like leaves, an obvious xerophytic adaptation. Gaya ribifolia is the plant of the drier mountains. It is rather more hairy than its close relation G. lyallii, and there are other differences, one being the “dripping point” to the leaf, at

times a most characteristic feature of this latter species. Although these distinctions are apparently trivial, they are sufficient to keep the two forms distinct from one another; and, although they may approach within a kilometre or two of one another, they are not found mixed together.^*

The two veronicas, V. traversii and V. leiophylla, are not especially xerophytic. It is true they are not so hygrophytic in form as Veronica salicifolia and its allies, but they show no very special adaptations for the conditions of life of a subalpine scrub. On the other hand, such conditions in the scrub under consideration do not seem to require very special adaptations, and the extreme xerophylly of Olearia coriacea is out of place. Even the other xerophytes could endure much severer conditions; and it seems here that we have another case of what I have pointed out several times in previous papers^†—viz., that such xerophytic structure is a survival from a former geological period when large areas of New Zealand were extremely arid.

Xerophytic structures are still more out of place on the floor of the subalpine scrub, and the list given above shows that such are hardly present. The only plant of special interest to be noted is the Ranunculus lobulatus. This was referred by Kirk to Ran. insignis as var. lobulatus, but he had never seen the flowers, and, as he showed, the leaves are very distinct from those of R. insignis. To my mind it is a well-marked species, and its affinities seem rather with R. munroi than with R. insignis. The leaves are large, with long but comparatively slender petioles. The laminæ are thin, reniform-orbicular, narrowly lobed, bright-green on the upper surface, pale beneath, variable in hairiness, but usually glabrous or nearly so on the upper surface, but with more or less very fine white hairs on the under surface and the petiole. Frequently the sinus at the base of the leaf is closed, and all sorts of transitions towards a peltate leaf ensue. The seedling has the first leaves quite entire, and with a cuneate base; the next leaves have an apical lobe; later leaves develop two lateral lobes, and the base becomes rounded and shows traces of a sinus. Evidently the young leaves are arrested stages of the adults. A large leaf may measure (lamina) 17 cm. by 15·5 cm., petiole 25 cm. There were a few flower-buds on some of the plants, and one had a flower nearly open. The sepals are broad, with a notched