Other formats

    TEI XML file   ePub eBook file  


    mail icontwitter iconBlogspot iconrss icon

Tuatara: Volume 9, Issue 1, September 1961

A Key to the Coprosmas of New Zealand — Part I

A Key to the Coprosmas of New Zealand — Part I

The genus Coprosma has its greatest concentration of species in New Zealand, where there are about forty of the ninety known species. The remainder are scattered through the Pacific, and fall within a line including Tasmania. Eastern Australia, Borneo, Hawaii, Juan Fernandez Islands and New Zealand's subantarctic islands. The genus was first collected by Banks and Solander in New Zealand on Cook's first voyage, but as their descriptions of seven species were never published the name valid to science originates from the Forsters, father and son, who came as botanists on Cook's second voyage. They found two species, one of which forced itself to their notice by its unpleasant smell, strongly reminiscent of carbon disulphide. As a result of this disagreeable first impression they stigmatised the whole genus with the name Coprosma meaning ‘smell of dung’, and applied the specific name foetidissima, ‘extremely vile smelling’, to the offending plant. Possibly this was uncharitable considering that only two species of the genus have a noticeably obnoxious smell, the New Zealand endemics C. foetidissima and C. crenulata. On the other hand, a friend of mine who used to live on Stewart Island remembers that she was taught to avoid mikimikis (Coprosma spp.) completely when collecting firewood, because even the seemingly inoffensive kinds were transformed into ‘stinkwoods’ by heat. The odour of page 32 C. foetidissima has been shown by Sutherland to be caused by traces of methyl mercaptan, and it would be interesting to know whether similar compounds occur in other Coprosma species, or in the related genus Nertera of which N. depressa at least smells like C. foetidissima.

Coprosma belongs to the family Rubiaceae, which includes such useful plants as Coffea and Cinchona, yielding coffee and quinine respectively; and Rubia, the madder, a source of red dye. Other New Zealand genera are Nertera and Galium. Nertera is close to Coprosma from which it is distinguished by its herbaceous habit and bisexual flowers. It grows in mats which would be hardly noticeable except that in season they bear a mass of shining orange-red fruits. A form with the popular name of ‘bead plant’ is sometimes cultivated in rockeries. Galium is a square-stemmed herb with small dry fruits consisting of two nut-like halves. An introduced Galium is well known for its ability to stick to clothes and skin by means of a covering of minute hooked hairs.

Coprosmas are all woody, but range in habit from creeping mat plants through different forms of shrub to small trees, and occur in all kinds of vegetation from the sea coast to the mountains. They have opposite pairs of leaves, interpetiolar stipules and pits or domatia can be found in all but the very small-leaved species. The stipules of Coprosma are not exactly equivalent to the paired lateral structures of the same name found at the base of the petiole in some alternate-leaved plants. In peas and beans, for instance, these are leaf-like, and could be thought of as similar in form and origin to the normal leaflets borne further out along the main axis of the leaf; but a Coprosma stipule bridges the gap between the bases of opposite leaves, and the two stipules and two leaf bases together encircle the node. In simple cases the stipules are more or less triangular, and may be either prominent or minute. They appear at every node on the plant including the cotyledonary node, in the form characteristic of the species. This interpetiolar type of stipule is probably best interpreted according to the theory of Sinnott and Bailey as a fusion of two formerly independent lateral stipules, since its venation is derived from the leaf traces on either side.

It is interesting to note that in Galium, the interpetiolar stipules are prominently expanded and leaf-like, with one or several occurring between the bases of each leaf pair, so that together the leaves and stipules form a whorl of similar organs at the node. So alike are the members of the whorl that the only reliable distinction between them is the presence of buds in the axils of the true leaves.

The stipules in Coprosma are associated with a complex collar surrounding the stem above the node and enclosing the two axillary buds. Fig. 1 illustrates the stipule collar of C. robusta cut away to show the nature of its parts. The two triangular stipules are joined to each other by a band of thinner tissue across the leaf axils and above the leaf abscission zones. This thinner tissue is the stipule page 33
Figure 1: Stucture of the stipule collar of Coprosma robusta. Drawing by Jean Perry. Figure 2: Coprosma robusta Raoul

Figure 1: Stucture of the stipule collar of Coprosma robusta. Drawing by Jean Perry.
Figure 2: Coprosma robusta Raoul

page 34 sheath. Besides being fused to each other in this manner the stipules are also joined edgewise to the leaf bases, and enclose a narrow space below the leaf abscission zones in which are the axillary buds. Or to explain it in another way, the leaves and stipules have been carried up on the rim of a cup by the growth of a ring of tissue at their bases. This is the stipule cup. C. robusta has both sheath and cup contributing to the stipule collar but in other species either one or the other may be developed. A deep stipule cup is often indicated by a band of green tissue persisting at the node when bark has been developed above and below. Any kind of stipule collar will at first constrict the axillary buds but later their growth hastens its rupture and loss.

Stipules probably always bear denticles at their upper free margin. These are stumpy or pointed outgrowths (literally ‘little teeth') which appear on close observation to be covered with a shining translucent jelly. Microscopic sections reveal the jelly to consist of elongated glandular cells packed into a dense palisade over a core of soft tissue. When the tips of the glandular cells break, their mucilaginous contents are released as a shining coating easily seen if young growing leaves are prised apart. In time this mucilage dries into a horny film and disappears. After the release of their contents the glandular cells shrivel while the internal tissue darkens and withers, reaching this stage at about the time the leaves of the same node are mature. Thus denticles, like the stipules which bear them, reach their fullest development at an early stage and their secretion probably protects the tender growing leaf tissues from desiccation. In C. serrulata and C. foetidissima the denticles are slim and could be regarded as multicellular hairs with a specialised glandular function. C. lucida and C. repens have similar but fatter, shorter denticles lacking the basal non-glandular portion of the slim ones. However, in C. robusta, C. tenuifolia, C. arborea and C. macrocarpa, the glandular region is the stipule apex itself, and although this is also usually referred to as a denticle, it is not quite the same structure as the hairlike denticles described above. C. australis stipules exhibit a particularly interesting intermediate condition, where the stipule apex is glandular as in C. robusta, but the glandular surface extends downwards and laterally to cover two rows of marginal projections which seem to correspond with the hairlike denticles of other species. The first pair of stipules on an axillary shoot is very compressed during development, and frequently the stipule nearest to the axis has a twinned apex. This is particularly noticeable in species with a simple glandular apex. e.g. C. robusta, C. macrocarpa and C. tenuifolia which have therefore been reported as having 1-2 stipular denticles. Twinned apices also occur in a similar position on axillary shoots of C. lucida and C. repens.

A Coprosma twig has a series of stipules from tip to base which vary in appearance from young to senescent. In some species the page 35
Figure 3: Coprosma australis (A. Rich.) Robinson Figure 4: Coprosma lucida J. R. et G. Forst.

Figure 3: Coprosma australis (A. Rich.) Robinson
Figure 4: Coprosma lucida J. R. et G. Forst.

page 36 stipules are very persistent, in others they begin to dry out and shrivel once the leaves have expanded to full size; but in all cases, wood formation within the main stem will increase its girth and subject the stipules to expansion and considerable lateral stretching, altering their shape before they are finally lost. It is best to take as characteristic examples those accompanying the youngest full-sized leaves.

The leaf pits or domatia in Coprosma are shared by its relative Coffea. They occur on the underside of the leaf in the acute angle made by the midrib at its junction with major secondary veins. Their sizes and shapes include variations from wide-mouthed shallow depressions to enlarged cavities entered by a tiny pinhole. They usually produce a raised blister on the upper surface of the blade. Domatia associated with secondary veins have been reported by Shirley and Lambert as occurring very occasionally in C. repens, and have also been observed by the writer on larger shade leaves not only in C. repens but also in C. australis and C. foetidissima.

The function of domatia is still a matter for speculation. Those who hold that they are shelters for small insects would call them domatia, meaning ‘dwellings’, and point out that other genera of Rubiaceae provide domatia for ants, Myrmecodia in a corky basal tuber, and Duroia in hollow stems or leaf swellings. Certainly small insects are often found in the pits, but they could have come upon this convenient shelter by accident. Others suggest that the occurrence of bacterial nodules in the leaves of Psychotria may give a clue to the nature of Coprosma pits. The bacteria of the nodules are beneficial because they fix atmospheric nitrogen, and are carried by the host from one generation to another by means of the seed, whence they invade the apical bud of the new plant and live in a gummy secretion inside the stipular sheath, infecting each new leaf as it develops. This recalls the secretions from Coprosma denticles and led Stevenson to investigate their bacterial content. She reported that the glandular cells as well as the secretion are full of bacteria and taking this together with other evidence that C. robusta at least is able to assimilate nitrogen from the air, concluded that the bacteria might be the agents responsible. However, the demonstration is not yet absolutely conclusive, and the relationship of the bacteria with the pits is not clear. The whole question is probably best left open for further investigation.

The leaf of C. serrulata has a remarkable colourless border which contains near its inner edge a small vein running parallel to the margin, and whose free edge projects in regular crenulations. Apart from a few references to thickened margins in other species, Coprosma leaves in general seem to be regarded as having unspecialised margins. Detailed investigations of leaf margins show that there exists a variety of conditions, from borders very like those of C. serrulata to an absence of any specialised structure. Oliver mentioned that the leaf margin of C. crenulata approaches page 37
Figure. 5: Coprosma repens A. Rich. Figure. 6: Coprosma serrulata Hook, f. ex Buchan.

Figure. 5: Coprosma repens A. Rich.
Figure. 6: Coprosma serrulata Hook, f. ex Buchan.

page break C. serrulata, and placed them close together in his classification. Both C. crenulata and C. foetidissima are probably nearest to C. serrulata in respect of their colourless borders, despite having crenulations on a smaller scale; although in C. crenulata as Oliver pointed out, the crenulations are more conspicuously developed towards the leaf tip. C. robusta and C. tenuifolia leaves are also bordered, but in young leaves the outermost cells project in almost hairlike points, sometimes persisting unchanged in mature leaves or becoming rounded into crenulations. Most of the other large-leaved species have a smooth, colourless border although in some it is narrow enough to be overlooked. Where the border itself is of even width, the leaf margin is sometimes irregular because it is slightly indented where the smallest veins run out to the edge. Small-leaved species of Coprosma may have a length of colourless border towards the leaf apex, or it is lacking altogether. The series described suggests the leaf margin of C. serrulata is the most complicated in the genus, but does not give support to the view that it stands definitely apart from all other species in this feature.

Male and female flowers of Coprosma are normally produced on separate plants, and they are in both sexes greenish and inconspicuous, being adapted for wind pollination. The stamens of the male have long threadlike filaments on which the anthers are suspended so delicately that the slightest air movement shakes out the pollen, and the styles of the female are two narrow divergent prongs which expose a considerable area of receptive surface. The dioecious condition associated with wind pollination ensures outcrossing and also favours hybridisation.

The fruits are the most attractive part of the plant. They range from delicate pearly white through bright shades of blue, yellow, orange, red and crimson to black. Cockayne regarded C. brunnea with its translucent blue fruits hanging like beads from the brown wiry stems, as ‘an élite of the flora’ for ornamental garden work. Their colouring appears to depend on the independent occurrence of two different classes of pigment. The fat-soluble carotenoid pigments giving yellow and orange-red colouring are found in the fruit flesh, but the water-soluble anthocyanins of the skin give blue, blue-black or crimson colouring. The fruits are popularly known as berries but are not berries in the strict botanical sense. They are drupes, closely allied to plums and peaches, because each of the two seeds contained in the fruit is enclosed by a hard covering layer, the stone. The seeds are dispersed by birds, passing unharmed through their digestive processes to be planted later in an ideally fertile environment.

The present section of the key deals with large-leaved species of Coprosma found in the main islands of New Zealand and depends entirely on vegetative characters, chiefly those of stipules. The accompanying plates depict the shoot tip of each species showing page 39
Figure 7: Coprosma macrocarpa Cheesem. Figure 8: Coprosma tenuifolia Cheesem.

Figure 7: Coprosma macrocarpa Cheesem.
Figure 8: Coprosma tenuifolia Cheesem.

page 40 stipules at the young stages which must be examined when using the key. A pocket magnifier is indispensable for accurate observation. In all the species plates, stippling represents dark or reddish pigmentation while parallel lines are used for shading. The specimens are arranged to show one of the stipules in face view and a pair in lateral view, with the associated leaf removed to show structures in its axil.

Key to Species

Plant with leaves generally over 25 mm. (one inch) long SECTION A
Plant with leaves about or less than 25 mm. long SECTION B
(To appear in Part II)

Section A Large-Leaved Coprosmas

1 Plant has a strong unpleasant smell when broken or bruised (Fig. 10). foetidissima
Plant without a strong unpleasant smell. — 2
2 Glandular surface of stipules not restricted to denticles but spreading well below them to form a clear diagonal boundary when observed from the side (Fig. 3). australis
Glandular surface of stipules restricted to the denticles. — 3
3 Stipules bear numerous glandular denticles which completely surround the stem. When a leaf is removed they can be seen in the axil. — 4
Stipules bear 1 to 3 (occasionally 5) glandular denticles grouped at the stipule apex. When a leaf is removed no denticles can be seen in the axil. — 5
4 Stipule sheath extremely narrow or lacking. Glandular denticles are plump and sessile. Leaf margin smooth (Fig 5). repens
Stipules have a deep axillary sheath. Glandular denticles are slim and hairlike, with the glandular surface restricted to the distal portion. Leaf margin has a conspicuous colourless crenulate border (Fig. 6). serrulata
5 Stipule hairless. — 6
Stipule hairy on surface or margin. — 7
6 Leaf margin has a colourless border fringed with fine hairs, observed most readily in young leaves. Stipules wither and break up irregularly in decay (Fig. 2). robusta
Leaf margin smooth or very slightly irregular, and hairless. Stipules become paper-thin with age, and are finally cut off neatly at a transverse abscission zone (Fig. 7). macrocarpa
page 41
Figure 9: Coprosma arborea Kirk Figure 10: Coprosma foetidissima J. R. et G. Forst.

Figure 9: Coprosma arborea Kirk
Figure 10: Coprosma foetidissima J. R. et G. Forst.

page 42
7 Stipule hairs are longer than the glandular denticle and surround it in a white tuft. Stipules become paper-thin with age, and are finally cut off neatly at a curved transverse abscission zone (Fig. 8). tenuifolia
Stipule hairs are shorter than the glandular denticle(s) and form a marginal fringe. Stipules wither and break up irregularly in decay. — 8
8 Leaf blade narrows abruptly then gradually towards the base to form a tapered wing on either side of the petiole. Leaf margin sometimes undulate. The leaf is tipped by a group of short furry hairs, often carried on a small apical projection. — 9
Leaf blade narrows evenly and the petiole lacks a wing. Leaf margin plane. * Leaf tip hairless. — 10
9 Young stipules and nodes darkly pigmented. Midrib feels ridged on both surfaces of the leaf. Stipule bears a single glandular denticle and hairs usually occur on the face as well as the margin of the stipule (Fig. 9). arborea
Young stipules have a transverse greenish band about the level of the leaf insertion which contrasts strongly with the dark colour of the stem below. Midrib feels ridged on the upper surface of the leaf only. Stipule bears 3-5 slim denticles on which the glandular surface is restricted to the distal part, and hairs occur on the margin of the stipule only (Fig. 10). foetidissima
10 Midrib feels ridged on the lower surface of the leaf only. Leaf margin has a colourless border fringed with fine hairs, observed most readily in young leaves. Stipule has an axillary sheath and a single glandular denticle. Members of a stipule pair diverge from the central stem when observed from the side (Fig. 2). robusta
Midrib feels ridged on both surfaces of the leaf. Leaf margin has a narrow colourless border lacking hairs. Stipule has no axillary sheath and bears usually 3 glandular denticles. Members of a stipule pair are strongly incurved when observed from the side and the denticles are close to the central stem (Fig. 4). lucida

Subscribers who intend to bind copies of Tuatara are advised that an index to the first ten volumes will be issued with Vol. X, No. 3.

page breakpage break

* Some specimens of C. lucida have a few delicate deciduous hairs at the leaf tip. As they disappear very early, and are unlikely to be detected without the aid of a powerful lens or microscope, C. lucida should key out satisfactorily under 10. Similar apical hairs sometimes occur on leaves of C. repens.