Art XXIII — New Species of Plants

Transactions and Proceedings of the Royal Society of New Zealand, Volume 48, 1915, Page 210

Art XXIII — New Species of Plants By T. F. Cheeseman, F.L.S., F.Z.S, Cuiator of the Auckland Museum. [Read before the Auckland Institute, 8th December, 1915.] 1. Geum divergens Cheesem. n. sp. Species distincta G. unifloro Buch. similis, sed scapis 1–4-floris, bracteis numerosis incisis, floribus flavis. Herba parvula, 4–10 cm. alta. Rhizoma breve, crassum, procumbens. Folia radicalia, lyrato-pinnata, 2–5 cm. longa; foliolis lateralibus minutis, terminali maximo orbiculari-cordato irregulariter dentato. Scapi 3.5–5 cm. longi; bracteis 2–5, lineari-lanceolatis, incisis. Flores 1–5, ratione plantae magni, 1.5–2.5 cm. diam, flavi. Calycis segmenta ovato-lanceolata, acuta vel acuminata. Petala magna, obovata, obtusa. Achenia dense villosa; stylis glabris, apice uncinatis. Hab —South Island: Sheltered places among rocks on the slopes of Mount Captain, Clarence Valley, alt 5,000 ft, T. F. C. Short, stout, 2–5 in high. Rhizome creeping, short, thick and woody, clothed with the bases of the old leaves. Leaves all radical, 1–2 in. long including the petiole, lyrate-pinnate; terminal leaflet very large, ¾–1 ¼ in. diam., orbicular reniform, indistinctly lobed, coarsely crenate-dentate; margins densely ciliate with long white hairs; upper surface sparsely villous, lower almost glabrous; lateral leaflets 2–4 pairs, minute, lanceolate or ovate-lanceolate, ciliate. Scapes 1 ½–3 m. high, slender, densely pubescent; bracts 2–5, the lowest sometimes ¾ m. long, lanceolate, incised or rarely inciso-pinnatifid. Flowers 1–5, large for the size of the plant, ¾–1 ¼ m. diam, yellow. Calyx-lobes ovate-lanceolate, acute or acuminate, sometimes with small accessory lobes at the base, pubescent. Petals much larger, obovate, obtuse. Achenes villous with long hairs, gradually narrowed into a long glabrous style hooked at the tip. Although this has something of the habit and appearance of G. uniflorum, it is by no means closely allied to that plant. The rhizome is shorter and not so stout, the leaves are thinner, with a different indumentum; the scapes are furnished with more numerous and much larger and more deeply incised bracts; the flowers are more numerous, sometimes as many as 5 to a scape, and are bright yellow, whereas they are solitary and always white in G. uniflorum. In size, in the shape of the leaves, and in the colour of the flowers it appears to approach the somewhat vague description given by Buchanan of his G. alpinum (a name already occupied). But he describes the flowers of his plant as “minute” and “⅕ m. diam,” a difference in size so great as to preclude the specific identity of the two plants, whatever “G. alpinum” may prove to be. 2. Olearia insignis Hook. f. var. minor Cheesem. n. var. Differt a typo habitu multo graciliore, foliis minoribus et angustioribus, pedunculis gracilioribus, capitulis ininoribus. Hab —South Island: Marlborough, between Kaikoura and Blenheim, H. J. Matthews!

Much smaller and more slender than the type; branches much less robust and less thickly clothed with tomentum. Leaves 2 ½–4 in. long including the petiole, 1–1 ½ in. broad, narrow elliptic-oblong to oblong-spathulate, gradually narrowed at the base into a rather slender petiole 1 in. long, much thinner and less coriaceous than in the type, tomentum on the under-surface much less dense. Peduncles 2 or 3 towards the ends of the branches, 3–5 in. long, slender, angular, sparingly tomentose, furnished above the middle with 2 or 3 linear-spathulate bracts. Heads small, ¾–1 ¼ in. diam., longer and narrower in proportion than those of the type, almost campanulate in shape; involucral bracts narrower and less densely tomentose. So far as can be judged from the rather scanty suite of specimens in my possession, this is a well-marked variety, distinguishable from the typical form at a glance by its much smaller size, more slender habit, less copious tomentum, much more slender peduncles, and smaller heads. It appears to keep its characters in cultivation. 3. Celmisia Thomsoni Cheesem. n. sp. Ab. C. bellidioides Hook. f. differt caulibus brevioribus et densissime compactis, foliis magis coriaceis rosulatis et arcte imbricatis, superne distincte setulosis. Herba pusilla, caespitosa. Rhizoma breve, ramosum, prostratum; ramis numerosis, brevibus, densissime compactis, 2–5 cm. longis, superne foliosis. Folia numerosa, imbricata, rosulatim disposita, 0.75–1.25 cm. longa, 3.5 mm. lata, anguste obovato-oblonga aut lineari-oblonga, obtusa vel obtuse apiculata, basi attenuata, coriacea, supra copiose et distincte setulosa, marginibus minute denticulatis; petiolis brevibus. Scapi graciles, 4–7 cm. longi, 2–4-bracteati, glabra aut sparse glanduloso-tomentosa; bracteis angustis. Capitula. 1–2 cm. diam.; involucri squamae paucae. Achenium sericeum. Hab. South Island. Eyre Mountains, Central Otago, in rock-crevices on the faces of cliffs, alt. 5,000–6,000 ft., W. A. Thomson and J. Speden! Forming compact cushion-like patches 1–3 in. diam., rarely more. Root-stock short, much branched, prostrate; branches usually numerous, densely compacted, leafy above. Leaves numerous, closely imbricate and rosulate, ⅓–½ in. long, ⅛–⅕ in. broad, narrow obovate-oblong or almost linear-oblong, obtuse or bluntly apiculate, narrowed at the base, thickly coriaceous, dull green, glabrous beneath, upper surface setulose with short stiff glandular hairs; margins furnished with minute irregular denticles; petioles short, broad, sheathing at the base. Scapes from near the tips of the branches, 1 ½–2 ¾ in. long, very slender, glabrous or more commonly sparsely glandular-tomentose above; bracts 2–4, narrow-linear. Heads ½–¾ in. diam.; involucral bracts few, narrow linear-oblong, acute, green or the outer ones purplish-green. Rays numerous, spreading. Achenes densely silky. I have much pleasure in dedicating this interesting plant to Mr. W. A. Thomson, of Dunedin, who in company with Mr. J. Speden was the first to collect it. Although allied to C. bellidioides, it differs from that species in several important characters. It is much smaller and much more compactly branched, thus assuming a tufted cushion-like habit quite different from the creeping and mat-like appearance of C. bellidioides. The leaves are much more closely placed, and are decidedly rosulate, while the upper surface is sprinkled all over with short stiff glandular hairs. The scape

is also much more slender than in C. bellidioides. When fresh the leaves are of a dull green, quite unlike the bright shining green of C. bellidioides. Both Mr. Thomson and Mr. Speden inform me that it usually occurs in crevices on the shaded sides of rocky cliffs; often, in fact, in situations where no direct rain can fall upon it, although drifting snow or fog might reach it. So far as my experience goes, C. bellidioides is usually found draping wet cliffs or creeping over wet gravelly slopes. 4. Cotula Willcoxii Cheesem. n. sp. Species C. pectinatae Hook. f. affinis, sed foliis non pectinato-pinnati-fidis facillime distinguenda. Caules 3–15 cm. longi, prostrati, copiose ramosi, radices fibrosas emittentes, sparse sericeo-pilosi. Folia petiolata, cum petrolo 0.5–1.25 cm. longa, glabra vel parce pubescentia, pinnatisecta; lobis 1–2-jugis, lineari-oblongis, subacutis. Pedunculi gracillimi, nudi, sparse pilosi, 2–5 cm. longi, foliis multo longiores. Capitula 0.75–1 cm. diam.; involucri squamae 1–2 seriales, oblongae, obtusae, inarginibus scariosis et denticulatis apice purpureis. Flores foeminei 2–3 seriales; flores disci tubuloso-infundibuliformes, 4-dentati. Achenia anguste obovata. Hab.—South Island: Head of Lake Wakatipu, Otago, near Mount Earnslaw, W. Willcox! Stems 1 ½–6 in. long, creeping and rooting, copiously branched and usually forming compact patches, sparsely silky-pilose or sometimes nearly glabrous. Leaves alternate, petiolate, with the petiole ⅕–½ in. long, glabrous or faintly pubescent, gland-dotted, pinnatisect; segments usually a single pair with a terminal one, rarely two pairs, flat, linear-oblong, subacute, usually quite entire or very rarely one or both of the lower segments are forked at the tip. Peduncles long, slender, naked, sparingly pilose, ¾–2 in. long, much exceeding the leaves. Heads ¼–⅓ m. diam; involucral bracts in 1–2 series, oblong, obtuse, purplish at the tips, margins scarious, jagged. Female florets in 2–3 sries, florets of the disc narrow funnel-shaped, 4-toothed. Achene narrow-obovoid. A well-marked species, closely allied to C. pectinata in the structure of the flower-heads, but in other respects presenting well-marked points of difference. It is larger, and much more glabrate, the peduncles are longer and more slender; and the leaves are never pectinate-pinnatifid, as is always the case in C. pectinata. Usually the leaves have only a single pair of segments with a terminal one, but more rarely there are two pairs of segments, in that case resembling some small forms of C. pyrethrifolia. From that plant, however, it is at once distinguished by the bisexual flower-heads, to say nothing of other differences. The only specimens that I have seen are those collected by Mr. Willcox, of Queenstown, and I have much pleasure in associating his name with the plant. 5. Senecio lapidosus Cheesem. n. sp. Affinis S. Monioi Hook. f., sed minore, et capitulis solitariis maxime differt. Fruticulus humilis, prostratus vel suberectus, densissime ramosus, 15–18 cm. altus, 20–25 cm. diam. Rami numerosi, divaricati et densissime compacti, basi robusti, lignosi, cortice pallido; ramulis apicibus ascendentibus, foliosis, lanato-tomentosis. Folia 2–4 cm. longa, 1–1.5 cm. lata, oblonga vel elliptico-ovata, obtusa, coriacea, supra arachnoideo-tomentosa,

subtus dense et appresse argenteo-tomentosa, marginibus minuti crenatis. Pedunculi e summis ramulis, solitarii, graciles, 1.5–2.5 cm. longi, dense argenteo-tomentosi; bracteis multis, linearibus vel lineari-oblongis. Capitulum solitarium, 1–1.5 cm. latum; bracteis involucri 12–15, linearibus, acutis, tomentosis. Flores radii 12–15, flavi; flores disci numerosi. Achenia numerosa, longitudinaliter sulcata, hispida. Hab.—South Island: Faces of rocky cliffs at Hell's Gate, Mason River, North Canterbury, H. J. Matthews! (1909). A small depressed much-branched prostrate or suberect shrub, forming dense clumps 6–8 in. high and 8–10 in. diameter. Branches many, closely compacted, stout and woody at the base, and covered with a pale flaky bark; branchlets ascending or suberect, leafy, densely woolly tomentose. Leaves ¾–1 ¼ in. long, ⅓–½ in. broad, narrow oblong or narrow elliptic-ovate, rarely oblong-spathulate, obtuse, coriaceous, margins minutely crenate, upper surface thinly covered with cobwebby tomentum, beneath clothed with dense silvery-white tomentum; petioles somewhat slender, broader at the base. Peduncles from the tips of the branchlets, stout or slender, 1–2 in. long, densely silvery tomentose; bracts numerous, linear or linear-oblong. Heads solitary, turbinate, ⅓–⅔ in. diam.; mvolucral bracts 12–15, linear, acute, densely tomentose. Ray-florets 12–15, yellow; disc-florets numerous. Achenes linear, longitudinally grooved, hispid. This interesting plant was collected by the late Mr. H. J. Matthews in 1909 in the gorge of the Mason River, North Canterbury, where it was associated with Epilobium brevipes, Veronica rupicola, and other species. So far, it has not been found elsewhere. It is clearly allied to S. Monroi, but is a much smaller and more compactly branched plant, and the peduncles are simple with a single terminal head, whereas in S. Monroi the inflorescence is copiously corymbosely branched. It was introduced into cultivation by Mr. Matthews, and forms an excellent plant for the rock-garden. I have to thank Mr. A. Bathgate for fresh flowering specimens from his garden in Dunedin, which are rather more slender than the wild specimens originally forwarded to me by Mr. Matthews. 6. Veronica obtusata Cheesem. n. sp. (V. macroura Hook. f. var. dubia Cheesem., Man. N.Z. Flora, 501). Species V. macrourae Hook. f. et V. diveigente. Cheesem. affinis, sed foliis bracteis et calycis segmentis inargine albo-pubescentibus differt. Frutex parvus, ramosus, 0.75–1.25 m. altus. Ramuli numerosi, patuli vel decumbentes, juniores teretes, pubescentes sed demum glabrescentes. Folia patula, sessilia vel brevissime petiolata, 2.5–5 cm. longa, 1.5–2.25 cm. lata, oblonga vel elliptico-oblonga vel obovato-oblonga, obtusa vel sub-acuta, subcoriacea, marginibus dense albo-pubescentibus. Racemi ramu-lorum apicem versus dispositi, 3.5–6 cm. longi, dense multiflori, recti vel curvati, rhachis cum pedicellis bracteisque dense pubescens. Calycis segmenta ovato-oblonga, acuta vel obtusa, marginibus conspicue albopubescentibus. Capsula 4 mm. longa, ovata, subacuta. Hab.—North Island: Sea-cliffs on the coast north of the Manukau Harbour, T. F. C. A branching shrub 2–3 ft. high; branches spreading or procumbent, the younger ones terete, at first pubescent, but at length becoming nearly glabrous. Leaves sessile or very shortly petiolate, 1–2 in. long, ⅗–1 in. broad, oblong to elliptic-oblong or obovate-oblong, obtuse or subacute,

subcoriaceous, flat, margins edged with a line of short and dense white hairs, midrib above occasionally downy. Racemes longer than the leaves, 1 ½–2 ½ in. long, straight or curved, dense-flowered, but not so much so as in V. macroura; rhachis and pedicels densely pubescent. Flowers ⅕ in. diam., pale bluish-white. Calyx 4-partite; segments oblong to ovate-oblong, obtuse or subacute, margins conspicuously ciliate with soft whitish hairs. Corolla-tube exceeding the calyx; limb 4-lobed. Capsule ⅕ in. long, ovate, compressed, not twice as long as the calyx. In the Manual I treated this as a variety of V. macroura. Since then I have obtained a much better series of typical V. macroura, mainly collected in the East Cape district by Bishop Williams, and find that the two plants are amply distinct. V. obtusata is much smaller, and has a much more diffuse mode of growth; the leaves are smaller, broader, and more obtuse, and the margins have a dense edging of short white hairs. The racemes are smaller and shorter in proportion, the flowers are not so dense and slightly larger; and the calyx-segments have their margins conspicuously ciliate with white hairs. It is probably nearer to my V. divergens, but that species has smaller and narrower glabrous leaves, longer racemes, and more glabrous calyx-segments. With respect to V. macroura, I have been informed by Mr. N. E. Brown, of the Kew Herbarium, that the localities of “Whangarei and Cook Strait,” given by Hooker in the Handbook, were probably quoted from Colenso's correspondence, as there are no specimens from thence in the Kew Herbarium. Mr. Brown further states that the only locality given on the type-sheet of V. macroura is “East Cape,” written against a specimen (Colenso, 101) by Sir J. D. Hooker. So far as I am aware, no one has gathered typical V. macroura at either. Whangarei or Cook Strait of late years. 7. Thelymitra pauciflora R. Br., Prodr. 314; Fitzgerald, “Australian Orchids,” vol. 1, part 6, t. 2. Stems slender, wiry, flexuous, 6–12 in high, rarely more. Leaf much shorter than the stem, narrow-linear, 1/10–⅕ in broad, rarely more, thick and fleshy, longitudinally grooved, deeply channelled in front, and thus concave Flowers 1–6, ⅓ in long, usually pale-blue or whitish-blue. Sepals and petals narrow ovate or ovate-lanceolate, acute. Column short, stout, the wing continued behind the anther and longer than it, 3-lobed, the middle lobe much the largest and also the highest, narrower than in T. longifolia, thick and swollen, projecting over the anther, deeply emarginate or 2-lobed, brownish-red at the base, bright yellow towards the tip, lateral lobes smaller at the base, projecting forwards almost horizontally, then suddenly bent upwards and erect, terminated by a dense brush of white cilia. Anther broad, connective produced into a short point. Hab—North Island. Leptospermum-clad hills in the. Auckland district, T. F. C, hills near. Pukekohe, W. Townson! vicinity of Kaitaia (Mongonui County), H. B. Matthews! Probably widely distributed in the Auckland Provincial District. Flowers from the middle of October to mid-November. I have been acquainted with this for many years, and have been accustomed to regard it as a variety of T. longifolia, to which species Mr. Bentham reduced Brown's T. pauciflora. But the structure of the column is very different from that of T. longifolia, and as long as the species of the

genus are principally founded on deviations in the form of that organ it is difficult to avoid the belief that the two plants are distinct. The late Mr. R. F. Fitzgerald, in his magnificent work on Australian orchids, unhesitatingly accepted this view; and a comparson of the figures of the two species given by him shows how great the differences are. The receipt of a large parcel of fresh specimens of T. pauciflora collected by by Mr. Town-son near Pukekohe gave me an opportunity of reviewing the matter, with the result of fully supporting the correctness of Mr. Fitzgerald's opinion. T. pauciflora differs from T. longifolia in its smaller size and more slender habit; in its narrower and deeply channelled leaf; in the smaller flowers; and especially in the middle lobe of the column-wing, which is deeply emarginate or 2-lobed, whereas it is much broader, more hood-shaped, and barely emarginate in T. longifolia. I should perhaps say that in all essential points New Zealand specimens agree with the drawing given by Mr. Fitzgerald of T. pauciflora. Hooker's T. Colensoi, which has not been collected of late years, and which was originally referred to T. pauciflora, differs from that plant, according to Hooker, “in the very narrow sepals and petals, very short column, and very long erect appendages.” 8. Rhopalostylis Cheesemanii Beccari in Herb. Kew., MS. In August, 1887, I accompanied the expedition sent by the New Zealand Government to annex the Kermadec Islands to the Colony of New Zealand. During this visit I was able to give rather more than a week to the exploration of the flora of Sunday (or Raoul) Island, and to form a fairly complete collection of the phaenogamic plants and vascular cryptogams of the island. Included in this series was a palm which I then supposed to be identical with the Norfolk Island Kentia Baueri (now known as Rhopalostylis Baueri), although, as no actual comparison of specimens could be made, an element of doubt still remained. A short time ago, however, the veteran botanist Dr. O. Beccari, one of the chief authorities on palms, compared my specimens with the Norfolk Island species, and informs me that the Kermadec plant is specifically distinct. He states that “it is closely allied to R. Baueti, but is easily distinguishable by its larger spherical fruits and larger leaves with considerably longer leaflets.” I hope that Dr. Beccari will shortly publish a full diagnosis; but in the meantime it is well to publish the fact that the Kermadec plant is a distinct and endemic species, and that R. Baueri does not extend beyond the confines of Norfolk Island.