Flora of New Zealand Series

Type: W, W. R. B. Oliver, Jan. 1929.

"A shrub, 2-3 meters tall, with slender, erect, opposite branches arising at narrow angles from the stem. Branchlets slender, green, glabrous." Bark of branches reddish brown; branchlets sts whorled. Lvs on slender petioles 10-20 mm. long. Stipules triangular, denticle prominent. Lamina bright green above, not shining, paler below, thinly coriac., narrow-obovate to obovate, acute to obtuse, midrib mucronulately produced; gradually narrowed to petiole; (65)-100-110-(120) × (20)-30-35 mm.; margins waved to subundulate. Reticulated nerves ± evident on both surfaces. ♂ not observed. ♀ in clusters of 3 on peduncles 10-20 mm. long; calyx-teeth rather long, acute; corolla not seen. Drupe orange-red, oblong, 6-9 mm. long.

DIST.: Great Barrier Id. "It is found only as an undershrub in the moist forest on the summit of Mount Hobson, altitude 2000 feet."

HYBRIDISM

Natural hybridization is now known to occur freely in the genus in N.Z. and to explain a great deal of the polymorphy shown. Cockayne (New Phytol. 22, `1923, 113) placed C. cunninghamii Hook. f. as one form of the hybrid group propinqua × robusta. He listed also banksii × colensoi and with doubt cuneata × depressa. Allan (Genetica 8, 1926, 155-160; 11, 1929, 335-346; T.R.S.N.Z. 68, 1938, 34-36) described the progeny arising from his artificial cross propinqua female × robusta male, up to the fruiting stage of the F2 plants and illustrated the If-forms and fr.-colours. Allan (Genetica 9, 1927, 511-515) described and illustrated numerous forms in a particular population where C. banksii, C. colensoi and C. foetidissima were present, and decided that the 3 spp. were all concerned in the hybrid progeny, the smell of foetidissima being traced even in specimens very closely resembling banksii. Cockayne (Acta for. fenn. 34, 1929, 21) listed hybrid groups involving 10 forest-inhabiting spp. Cockayne and Allan (Ann. Bot., Lond. 48, 1934, 43) listed 23 hybrid groups (2 with doubt).

Oliver (loc. cit. 4-17, 179-196) goes into the question in some detail, and his discussion should be consulted by the student. He accepts the following groups: acerosa × repens (C. kirkii Cheesem.); arborea × spathulata; australis × robusta; australis × tenuifolia; banksii × colensoi; banksii × foetidissima; colensoi × foetdissima; crassifolia × rigida; crassifolia × robusta (C. buchananii Kirk); crassifolia × tenuicaulis; foetidissima × rugosa (from specimens now in A collected by Cranwell and Moore on Maungapohatu, "Apparently a hybrid of which one parent is Coprosma rugosa; the other may be C. foetidissima"); lucida × rhamnoides (C. gracilis A. Cunn., C. viridis Carse); lucida × robusta; obconica × propinqua ("Wairoa Gorge, Nelson, 1913, Gibbs, Brisbane Herb. Apparently it is a hybrid between C. obconica and another species probably C. propinqua"); parviflora × robusta; propinqua × robusta (C. cunninghamii Hook. f.); propinqua × repens; repens × rhamnoides (C. neglecta Cheesem.); rotundifolia × tenuicaulis (× C. gracilicaulis Carse). Oliver lists a further 13 groups recorded by Cockayne and Allan, jointly or separately, but in the absence of supporting specimens does not discuss them. I use the names adopted in the text; those marked D certainly need confirmation: areolata × rotundifolia; astonii × colensoi; astonii × foetidissima, D; australi s × lucida, D; cheesemanii × depressa, D; cheesemanii × pseudocuneata, D; foetidissima × lucida, D; foetidissima × pseudocuneata; parviflora × depressa, D; parviflora × propinqua; parviflora × rigida; pseudocuneata × depressa, D; rhamnoides × rotundifolia.

Specimen 8999 in A, in the C. kirkii folders, is a small branch with crowded ovate to broad-oblong lvs ± 7-25 mm. long, with a note by Cheeseman "Coprosma raised from seed by Bp. Williams in his garden at Taumate, Napier. Seed taken from a ♀ of C. Kirkii, originally from Portland Island. Probably a cross between C. Kirkii and C. baueri. Sept. 1912."

I have seen only herbarium material. The citations are from Oliver, loc. cit. 127.

DOMATIA

Cheeseman (T.N.Z.I. 19, 1887, 221) noted: "In nearly all the species, except a few of the smaller-leaved ones, curious little pits exist on the under-surface of the leaves, in the axils formed by the union of the primary veins with the midrib. They are never more than ⅓ inch in length, and are usually much less. Inside they are lined with numerous stiff white hairs . . . They are often inhabited by a minute yellow acarid". Greensill (T.N.Z.I. 35, 1903, 342-355) describes and illustrates these pits for 9 spp. and for × C. cunninghamii, but found no acarids present. Oliver (loc. cit. passim) briefly refers to their occurrence; they appear to have no taxonomic importance. Stevenson (Ann. Bot., Lond. 17, 1953, 343) says: "these pits are ringed round or partly closed with hairs similar to those found on the stipules or elsewhere on the plant. The young hairs of the domatia and those on the stipules and other parts are densely filled with bacteria. The young buds are also covered with a gum containing bacteria. In all cases a red pigment was seen to be present at some stage either in the cells which contain bacteria or in neighbouring ones".

STIPULAR DENTICLES

Cheeseman (T.N.Z.I. 19, 1887, 222) observes: "All the species possess interpetiolar stipules. They are more or less triangular in shape, often with minute denticulations towards the apex . . . At the apex of the very young stipule a gland is situated which secretes a copious supply of a viscid mucilaginous fluid. These glands are highly developed and in an active state when the adjacent leaves are in the early stages of growth, but shrivel up and cease to secrete long before the leaves attain their full size." Oliver (loc. cit. passim) makes frequent reference to these deticulations, and they appear to be of definite taxonomic value. Stevenson (loc. cit. 343) remarks: "On all the large-leaved species, bacterial nodules are conspicuous on the stipules". Of the smaller-leaved spp. 13 are listed as having "one or more slender less conspicuous, stipular nodules".

SEXUAL EXPRESSION

Cheeseman (T.N.Z.I. 19, 1887, 223) noted that "The flowers are unisexual, and the sexes are placed on different plants. Occasionally, however, a few male flowers are intermixed with the female, and vice versa. Some species, and especially C. robusta and C. foetidissima, now and then produce hermaphrodite flowers, to all appearance well-developed and perfect, but which seldom mature fruit." Colenso (T. N. Z. I. 21, 1889, 85) described for his C. pendula ♂,♀ and hermaphrodite fls, all "on one branchlet". Of his C. coffaeoides (loc. cit. 88) he says, "Specimens of the male plant have been seen carrying female flowers at top of branchlet."

Wild and Zotov (T.N.Z.I. 60, 1930, 547-555) record their observations "on the incidence of abnormal, irregular, and the hermaphrodite flowers . . . assuming always that the normal plant is dioecious." Three useful figures are given. C. acerosa, areolata, propinqua were strictly dioec. × C. cunninghamii showed some hermaphrodite fls, these confined to male plants. Comparatively slight irregularities were found in depressa, repens, rigida, rotundifolia. The situation was more complex in australis, foetidissima, lucida, rhamnoides, robusta, tenuifolia. The paper should be consulted for details. There remains much work to be done.

POLYMORPHY AND PLASTICITY

Cockayne (T.N.Z.I. 38, 1906, 341-344) discussed and illustrated a case of If-variation and its causes in C. repens. He recorded (T.N.Z.I. 39, 1907, 378-379) that a prostrate plant of C. propinqua from Chatham Id, cultivated in his Christchurch garden, remained prostrate over several years. Oliver (loc. cit. passim) discusses polymorphy and plasticity for most of the N.Z. spp., and should be consulted for details.

CHEMICAL ASPECTS

Aston (N.Z. J. Sci. Tech. 1, 1918, 346-347) in his study of Coprosma as a source of dyes treated the cortex of the following with a 10% solution of sodium hydroxide; the untreated and treated colours are given in sequence: C. acerosa ―yellowish green, brownish red; × Kirkii ―white to yellowish, pinkish red to crimson on yellowish cortex only; repens―white, nil. C. propinqua ―white, greenish yellow; × cunninghamii ―white, greenish yellow; robusta―white, greenish brown; × buchananii ―greenish white, nil; crassifolia ―white, nil. C. linariifolia ―yellowish, red; microcarpa ―yellowish green, reddish brown to brick red; parviflora ―white, nil; colensoi ―greenish, pinkish red; foetidissima ―greenish, yellow; pseudocuneata ―brownish, red; tenuicaulis ―yellow, blood red; areolata ―dark brown, violet; rhamnoides ―greenish yellow, orange; rotundifolia ―orange, light to dark red; australis ―light orange, port―wine red; lucida ―canary yellow, crimson.

Briggs and his co-workers (J. Chem. Soc. 1948-1954) have contributed 9 papers on the "Chemistry of the Coprosma Genus". They have studied in detail the colouring mattters from a number of spp. and find them to be anthroquinone derivatives, with rubiadin or its methyl ether occurring in most spp. as yet examined. The results, when completed, promise to be of considerable taxonomic interest. Briggs and Thomas (loc. cit. May 1949, 1247) point out that the anthroquinone compounds of C. acerosa are very similar to those of C. lucida, though these spp. are placed wide apart in current arrangements. Briggs and Nicholls (loc. cit. Nov. 1954, 3941) note that "The stem bark of Coprosma arborea Kirk, C. tenuifolia Cheesem., C. cuneata Hook. f., C. pumila Hook f., C. repens Rich., C. rhamnoides A. Cunn., and C. robusta Raoul is practically devoid of anthroquinone colouring matters but sucrose has been isolated from extracts of the two first-mentioned species."

INCERTAE SEDIS

1. C. divaricata A. Cunn. in Ann. nat. Hist. 2, 1839, 207 is diagnosed thus: " foliis obovato-oblongis obtusis petiolatis glabris, floribus axillaribus solitariis breviter pedunculatis, ramis divaricatis rigidis, ramulis tomentosis. New Zealand (Northern Island). Dry woods at Wangaroa.- 1826, A. Cunningham.-1833, R. Cunningham." Hooker (Fl. N.Z. 1, 1853, 107) accepts the sp., commenting: "What I take to be the ordinary form of C. divaricata resembles C. rhamnoides, but has large, generally more coriaceous leaves, less pubescent branches and larger berries." The type specimen, at K, has twiggy branches with a few suborbicular lvs left. Oliver (loc. cit. 90) places the name as an absolute synonym of C. rhamnoides. Four vars. are described by Hooker:

" Var. β. gracilis; ramis gracilibus, foliis lineari-spathulatis obtusis. C. gracilis, A. Cunn.  Herb. et Prodr. [2, 1839, 206] . . . In var. β the leaves are rather longer than in the ordinary form". Type: K, Bay of Islands, R. Cunningham, 1834. Oliver (loc. cit. 192) notes: "The type specimen of C. gracilis has slender, pubescent branchlets, thin, elliptic-ovate leaves, with pubescent petioles and small pubescent stipules" He places gracilis forms as the progeny of C. lucida × rhamnoides.

" Var. γ. pallida; ramis gracilibus, cortice albido, foliis obovato-spathulatis sublonge petiolatis . . . in var. γ the slender branches are covered with a white silvery bark". Type: K, Head of Wairarapa Valley, Colenso, 1847. Oliver (loc. cit. 95) places the name as an absolute synonym of C. virescens.

" Var. d. latifolia; ramis gracilibus, cortice fusco, foliis sparsis breve petiolatis majoribus (3/4 unc.) late oblongo-spathulatis apiculatis retusisve subtus pallidis glaberrimis v. puberulis. (An sp. distincta?) . . . in var. d the leaves are very broad, on short petioles, are 3/4 inch long, blunt, apiculate, pale and sometimes pubescent below". Type: K, Pahawa River, Colenso 1976. Noted on sheet as C. filiformis Col. n. sp. Branchlets slender, dark; lvs c. 1·75 × 1.1 cm.; fls c. 3 mm. long. There are 2 pieces, a portion of the lower one sent to W. Oliver (loc. cit. 83) considers the var. to be C. rubra Petrie.

" Var. ε. coriacea; ramis robustis lignosis, foliis sparsis rigide coriaceis late obovato-rotundatis spathulatisve marginibus recurvis glaberrimis . . . in var. ε the branches are very woody and thick, leaves smaller than in var. δ, but very coriaceous." Type: K, Manukau Bay, January 1843, Colenso 26. This, as Oliver (loc. cit. 82) decides, seems certainly to belong to C. crassifolia.

2. C. arcuata Col. in Tasm. J. nat. Sci. 2, 1846, 298. On clay hills in woods near Te Karaka (Hokianga Harbour) Colenso collected a "slender shrub, with long drooping filiform branches. Coprosma arcuata, foliis (parvis) obovato-oblongis sub-spathulatisve truncatis seu emarginatis basi attenuatis petiolatis glabris subfasciculatis, margine incrassata; ramis valde arcuatis dependentibus, ramulosis villosis; caulis arbusculus sesquiorygialis gracilis." Cheeseman (Man. N.Z. Fl. 1906, 1141) considered C. arcuata to be "probably the same as C. crassifolia" Colenso's description does not at all well agree with this view. I have not seen specimens; one of the F2 progeny of the artificial cross propinqua × robusta answers the description rather well. Oliver (loc. cit. 8) decided, "There are no specimens in Colenso's herbarium labelled C. arcuata. So the name is best abandoned as indeterminable."

3. C. solandri Kirk in T.N.Z.I. 29, 1897, 522 is thus described: "Much branched, branches stout, obscurely tetragonous, rigid; branchlets numerous, short, erect; bark whitish, setose. Leaves sessile, very coriaceous, linear-lanceolate, about ⅓ in. long, 1/10 in. broad, acute or apiculate, sparsely ciliate, laxly imbricating, erect, midrib sunk on both surfaces. Stipules setose, ciliate, loosely sheathing. Flowers not seen. Fruit solitary, terminal; involucel of two short leaves with dilated bases, globose-ovoid, 4/16 in. to 5/16 in. long, the persistent calyx-lobes acute, ciliate, connivent. East Cape district. Herb. Banks. Apparently related to C. linariifolia and C. Colensoi, but its affinities cannot be precisely indicated in the absence of flowers." Later (Stud. Fl. 1899, 242) he added: "Apparently a much-branched shrub. Branches as thick as a goose-quill . . . From the Banksian Herbarium, British Museum. This fine species is related to C. linariifolia and C. Colensoi, from both of which it differs in the stout tetragonous branchlets and sessile erect imbricating coriaceous leaves." Cheeseman (Man. N.Z. Fl. 1925, 872) accepts the sp., in the absence of the specimens used by Kirk; "There are specimens, however, in the set of Banksian plants presented to my own herbarium from the same source. Unfortunately they are not in flower, but the aspect of the plant appears to resemble that of C. linariifolia." Oliver remarks (loc. cit. 8): "The original description of C. Solandri does not exactly match any known species of Coprosma. It recalls both C. pseudocuneata and C. Cheesemanii; but as the specimens on which Kirk founded the species are lost, the name Solandri can be dropped as indeterminable."