Art. XXXVI.—On the Punui of Stewart Island, Aralia lyallii, n.s.

Transactions and Proceedings of the Royal Society of New Zealand, Volume 17, 1884, Page 293

Art. XXXVI.—On the Punui of Stewart Island, Aralia lyallii, n.s. By T. Kirk, F.L.S. [Read before the Wellington Philosophical Society, 9th July, 1884.] Plate XVII The Punui, one of the most striking plants in the New Zealand flora, was discovered by Dr. Lyall in 1848, and referred by Dr. Hooker** Fl. N.Z. i., p. 95. to Aralia polaris, Homb. and Jacq., discovered by those botanists on the Auckland Islands in 1839. Subsequently Dr. Hooker's suggestion that Aralia polaris might form a new genus to be termed Stilbocarpa was carried out by Decaisne and Planchon, and in the Handbook of the New Zealand Flora published in 1864, the Auckland and Stewart Island plants are included under the description of Stilbocarpa polaris, Decn. & Plan., the author stating that Lyall's plant wants the long bristles which are so characteristic of the Auckland Island plant, and might belong to another species†† Handbook N.Z. Flora, p. 100. For the sake of conciseness, in the following paragraphs the native name punui will be restricted to the Stewart Island plant; the generic name Stilbocarpa to that from the Auckland Islands. Until recently so little has been known of the flora of Stewart Island that, notwithstanding the botanical interest attached to such a remarkable plant as the punui the material for determining its precise relationship to Stilbocarpa has not been available: no specimens were to be found in our Herbaria, no plants in our botanic gardens. In 1878 a fine living specimen

with other Stewart Island plants was kindly placed at my disposal by Mr. C. Traill, who shipped them by the Government steamer “Stella,” which also brought several cases of plants from the Auckland Islands for the Colonial Botanic Gardens. As my case was by some mistake removed to the gardens, I allowed the plants to remain there, so that the punui was grown under the same conditions as the Stilbocarpa, all the plants at first being planted in the shade-house. While closely agreeing in the shape of the leaf, toothing, venation, etc., the two plants exhibited marked differences in minor characters—the texture of the leaf, amount of hairiness, and the colour and size of the hairs: the leaves of the punui being more membranous than those of the Stilbocarpa, the upper surface usually glabrous, the lower clothed with soft white hairs and the terete petiole. In Stilbocarpa both surfaces were clothed with fulvous bristly hairs, longer and stouter than those of the punui, the petiole was slightly compressed, and in some cases faintly grooved on the upper surface. In both plants the petiole was pilose. Two specimens of Stilbocarpa gradually developed new leaves in which the texture was less coriaceous, and the hairs reduced in number, shorter, and less bristly. These were exactly the results which I had observed in a greater degree with plants cultivated for two years in my own garden, so that the theory which attributed the trivial differences between the vegetative organs of the two plants to the different climatal conditions under which they grew, seemed to receive all the confirmation needed to establish their identity. Had good specimens of the flowers and fruit been available for examination, the opinion expressed by me** “Trans. N.Z. Inst.,” vol. xiv., p. 387. would have been different. Business, however, took me from Wellington before the punui had fairly developed flowers, and the single imperfect specimen that I was able to secure exhibited no characters calculated to alter my opinion. In 1880 Mr. J. B. Armstrong gave an imperfect description of the punui under the name of Stilbocarpa lyallii.†† “Trans. N.Z. Inst.,” xiii., p. 336. He states, “unfortunately I have not been able to obtain flowers or fruit, but there is no doubt as to the genus,” and describes the leaves as being “from 4 to 8 inches across or more, with a closed—not open—sinus.” As will be shown presently, both these statements are erroneous. Recently I have had the pleasure of examining the punui in its native island, and have received fully ripe fruit from Mr. C. Traill, the result being that not only must my old opinion as to its identity with Stilbocarpa polaris be abandoned, but that it must be removed to another genus.

Stilbocarpa was proposed by Sir Joseph Hooker as the name of a mono-typic genus, consisting of the Auckland Island plant already mentioned, and was established by Decaisne and Planchon in 1854, its essential characters being drawn from the 3–4-celled acetabuliform fruit, which in fact affords almost the only characters by which it can be separated from Aralia. The fruit of the punui, instead of exhibiting the cup-shaped cavity characteristic of Stilbocarpa, has a flattened apex covered with an epigynous disc consisting of the two stylopodia: in all respects agreeing with Aralia, to which it must consequently be removed, and in honour of its discoverer may worthily be designated A. lyallii. I now append a description. Aralia lyallii, n. s A stout herb. Stems ¾″ thick, pilose, forming strong arcuate stolons. Leaves alternate, crowded, petioles 1′–5′ long, fistulose, terete, pilose, with a sheathing laciniated ligule at the base: blade 6″–24″ in diameter, orbicular-reniform, lobed and deeply toothed, upper surface shining, usually glabrous, hairy beneath. Umbels monœcious, on axillary or terminal scapes, equalling or exceeding the leaves, globose 6″–12″ diameter, compound: primary involucral leaves foliaceous, inferior linear. Fl. unisexual, calyx teeth reduced to points, petals linear, more or less imbricate in bud. Male, stamens 5, filaments slender, disk 2-lobed. Female, stylopodia 2, reniform, forming a flat indented disk; styles 2, short, free, straight or divergent, ovary 2-celled. Fruit spherical, black, 2-celled, cells 1-seeded: testa crustaceous, striated. Hob. South Island:—Goal Island, Preservation Inlet (identified from the deck of a passing steamer); Stewart Island and outliers, chiefly on shady cliffs, etc; Herekopere Island, Ruapuke Island, Green Island, Centre, Island (nearly extinct). Reported also from the Snares, Antipodes Island, and Bounty Island, but I have not seen specimens. The punui often forms large patches spreading by means of the stout naked stolons which at first are suberect but gradually become inclined or arched until the terminal bud comes in contact with the ground, when roots are given off and a new plant is speedily developed. The stems vary in length from a few inches to 3′ or 4′ and are about the thickness of a man's finger. The patch becomes more and more dense as seeding plants are developed amongst the stolons. Specimens grown in the shade exhibit a marked difference from those grown in the open. In the former the leaves are flat or convex, more membranous, and with softer hairs than those grown in exposed places. The latter have leaves of stouter texture and clothed with stronger hairs—the blades often concave forming cups having the cordate or reniform bases folded inwards: it is doubtless this peculiarity

which led Mr. J. B. Armstrong to describe the leaf as having a closed sinus, an error which is at once demonstrated by simply flattening the leaf. As a rule large specimens growing in the open have the upper leaves concave and the lower convex, the texture of those most exposed being almost as thick as in some specimens of Stilbocarpa. Occasionally hairs are developed on both surfaces, but they are always white and soft. The foliaceous ligule at the base of each leaf is sometimes largely developed, and usually exhibits a pair of acute laciniæ larger than the others, at first sight presenting the appearance of a pair of ordinary stipules: most frequently all the laciniæ are more or less acute and ciliated, but a considerable amount of variation is exhibited in this respect. Much variation is shown in the density of the inflorescence; the umbels are highly compound, usually forming a dense orbicular mass of reddish-purple flowers, often more than a foot in diameter. Frequently the inflorescence is lax and open. The scapes are fistulose, and equal the petioles; the primary involucral leaves are sometimes 5″ or 6″ in diameter, on stout petioles, and closely resemble ordinary leaves; the secondary series is also petioled but greatly reduced in size and modified in form, being trifid, tripartite, or lobulate with a cuneate base; in the tertiary umbellules they are simply linear. In most specimens each of the secondary rays carries a female umbellule, from beneath which a whorl of from two to five male umbellules is given off; some of these are again branched. Sometimes the apex of a secondary ray terminates in a single abortive flower, when the tertiary rays carry small female umbellules surrounded by male as before. The pedicels of the male flowers are shorter than those of the female. In the male flowers the calyx teeth are reduced to points, in the female they are obsolete; the petals are usually close set, those of the male being slightly longer than the others, both alike are of a lurid reddish-purple. Stilbocarpa polaris differs from Aralia lyallii in the stouter texture of the leaves, which are clothed on both surfaces with long bristles, and the petiole is slightly compressed, sometimes exhibiting traces of a groove on the upper surface; the ligule is usually more deeply laciniated, the laciniæ being orbtuse and fringed with strong cilia. The petals are broader, obovate-spathulate, and of a pale yellow colour. The stamens are oblong, with rather shorter filaments, and the styles are recurved. The stylopodia form an interrupted annular disk, and as already shown, the fruit is acetabuliform 8–4 celled. It appears to be restricted to the Auckland Islands, Campbell Island, and Macquarrie Island.

Stilbocarpa Polaris, After Hooker Aralia Lyallii, T Kirk. T.Kirk, del

Explanation of Plate XVII I. Stilbocarpa polaris (after Hooker). 1. Bud. 2. Staminate flower with petals removed. 3. Stamen. 4. Pistillate flower. 5. Portion of ovary, showing a single ovule. 6. Section of ovary. 7. Ripe fruit. 8. Section of fruit. 9. Seed. 10. Longitudinal section of seed. 11. Embryo. II. Aralia lyallii, T. Kirk. 1. Bud. 2. Stamens. 3. Perfect flower. 4. Pistillate flower. 5. Longitudinal section of pistillate flower. 6. Transverse section of ovary. 7. Ovule. 8. Fruit. 9. Longitudinal section of seed.