Go to National Library of New Zealand Te Puna Mātauranga o Aotearoa
Volume 31, 1898
This text is also available in PDF
(4 MB) Opens in new window
– 354 –

Art. XXXII.—An Inquiry into the Seedling Forms of New Zealand Phanerogams and their Development.

[Read before the Philosophical Institute of Canterbury, 7th September, 1898, and 22nd February, 1899.]

Plates XXVII.–XXXIV.
Part I.: Introduction.

The investigation of seedling, forms of plants is a matter which, until quite recently, has been much neglected, and it was not until the year 1892 that this state of things was in some degree remedied by the publication of Sir John Lubbock's painstaking and admirable work entitled “A Contribution to our Knowledge of Seedlings.” In this treatise, which extends to 646 pages, are excellent drawings and descriptions of seedlings belonging to most of the natural orders, together with usually an account of the seed and the embryo. These descriptions in most instances treat of seedlings at certain stages of their growth, but do not trace the life-history of the plant from the germination of the seed to the completed form of the mature plant. The most elaborate portion of the work in question is the classification of the Cotyledons of each natural order, from which is deduced a theory to account for their various shapes. Concerning this theory I do not now propose to speak; any comment will be reserved until such time as my work is completed and its results summarised. Respecting seedling forms of our indigenous plants very little has been published. In the work just mentioned Linum monogynum, Forst., Pittosporum crassifolium, Banks

– 355 –

and Sol., Aristotelia racemosa, Hook, f., and Veronica salicifolia, Forst., are the only New Zealand species described. Kirk, in his various writings, has treated at some, length, especially in his “Forest Flora,” of the young forms of a number of our plants, but mainly those forms which occur together with the mature leaves on old plants, or young specimens of considerable size as found wild; in very few instances, however, has he described an actual seedling from the time of its germination.

From the above considerations it may be seen that a very considerable field is open for investigation by New Zealand botanists. That such investigation is of great interest and importance hardly needs asserting. It may be well, however, to show briefly some of the results which may follow such work. Taking, first of all, systematic botany, much light may be thrown on the relationship of forms, so that we shall, be able to determine much more accurately than has hitherto been possible the limits of many of our critical species. If, for instance, two plants have seedling forms exhibiting considerable differences, they would moat likely belong to different species; while should the two seedlings be identical it would be a strong argument in favour of the two being identical.

With regard to the first of the above hypotheses, Panax simplex, Forst., according to Kirk (“Forest Flora,” p. 212), seems a case to the contrary, since it is stated to have two distinct young forms “so widely different from each other that it is difficult to believe they can belong to the same plant.” In such a case as this a mistake may easily be made, by the most careful observer, and artificial investigation is necessary before the statement can for certainty be accepted. No one, unless well versed in seedling forms of New Zealand plants, however well he might know the mature form, could possibly recognise the seedlings of quite a number of our plants. Here, of the three figures on pl. cvi., only one—, No. 3—can claim to represent a young seedling, while Nos. 1 and 2 might very well be later developments of No. 3. Without, however, going into this subject in detail now, different environment, as will be shown later on, has a most marked effect on the same forms in many plants; hence this case, granted that the facts are as stated, may be merely a case of different development under different conditions. And this is the more likely since No. 3 is stated to be the only form found on the Auckland Islands (Hooker), and the prevalent form on Stewart Island (Kirk).

Besides the value of the study of seedlings from the specific point of view, that from the biological standpoint is of far more absorbing interest: the matter of varying environment, mentioned above, may lead to the discovery of most

– 356 –

important facts, while from the differences between the earlier and later developed leaves, and the younger and older forms of the plant, much information as to the phylogeny of a species or of an organ may be afforded. In his recent work, “A Text-book of Botany,” Strasburger says (page 46), “A highly organized plant which begins its development with the simplest stages and gradually advances to a higher state of differentiation repeats in its ontogeny its phylogenetic development”; and further on, “From the fossil remains of former geological periods it is safe to conclude that such Conifers as Thuja, Biota, and the various Junipers that now have scalelike compressed leaves have been derived from Conifers with needle-shaped leaves. This conclusion is further confirmed by the fact that on the young plants of the scaly Conifers typical needle-shaped leaves are at first developed. The modified leaf-form does not make its appearance until the young plant has attained a certain age, while in some Junipers needle - shaped leaves are retained throughout their whole existence.” Reference is also made to the Australian Acacias, whose early pinnate leaves in many cases become eventually phyllodes, and he concludes with these words : “That it is permissible on such phylogenetic grounds to conclude that the Australian Acacias have lost their leaf-blades in comparatively recent times, and have in their stead developed the much more resistant phyllodes as being better adapted to withstand the Australian climate. The appearance accordingly of the phyllodes at so late a stage in the ontogenetic development of the Acacias is in conformity with their recent origin.”

In a similar manner, Sir John Lubbock argues (“Flowers, Fruits, and Leaves,” page 141) “that the present furze is descended from ancestors with trifoliate leaves”; and further on he explains by an examination of seedling conditions how lobed leaves are of more recent origin than cordate leaves.

Besides the above-quoted instances, many others could, be cited from foreign plants,* but in no flora in the world is this phenomenon of such common occurrence as in the flora of New Zealand. Indeed, it will be seen from subsequent parts of this inquiry that a very large percentage of our plants exhibit most startling phenomena of this kind. And it is not

[Footnote] * See “Aspects of the Phænogamic Vegetation of Rodriguez,” by J. Bayley Balfour, D.Sc., F.L.S., in Journ. Linn. Soc. (Botany), vol. xvi., pp. 7–25.

[Footnote] † On this point I can speak with some authority, since, during the past few years I have personally raised from seed several thousands of species of extra-tropical plants, and in few, save certain Australian genera and Conifers, have I noticed any marked changes in leaf-form to take place.

– 357 –

the seedling alone which so behaves; but granted certain conditions—notably such as produce rapid young growth—and the mature plant will revert in part to the seedling form. Some species are so unstable in form that both forms may be found in a state of nature on the same plant at once—as, for instance, when one portion is exposed to sun and wind, while another portion is shaded. Thus at 1,500 m. on the Craigieburn Mountains I have found Veronica tetrasticha, Hook, f., with both its scale-like* and its-true (seedling) leaves, the latter occurring where a shoot was sheltered by a rock from sun and wind. Many other examples could be adduced from my own observations, but such will be reserved until dealing with each particular species.

Kirk was the first to point out such dimorphism in the whip-cord Veronicas (Trans. N.Z. Inst., vol. xi., p. 465), stating that “the dimorphism in the foliage of all the species characterized by appressed leaves has not received the attention it merits …. There can be little doubt that the free leaves are equally characteristic of the seedling state of the plant, although I have been unable to find them in a wild condition.” That was published in 1878, but quite recently, in the “Transactions of the New Zealand Institute” for 1895, volume xxviii., page 515, Kirk again refers to the subject, writing: “The species of the dimorphic—or, as it might with equal propriety be called, the mimetic—section are invested with special interest, the entire section, with the single exception of the Australian V. densifolia, being endemic in this colony. At present, strangely enough, our knowledge of the early leaves of these singular plants has been chiefly obtained from old specimens, on which they are often produced by reversion, especially under cultivation. The subject” [referring to the morphology of these species] “will not be satisfactorily worked out until the seedlings, as well as the more advanced stages, have been studied in a systematic manner.” Since such seedling leaves are usually developed on the mature plant—as already pointed out—by conditions which promote vigorous growth, can it be upheld, as a contrary theory to such leaves being ancestral forms, that they are merely special organs designed for the rapid growth of the plant in its early and most insecure period of existence among its often most inhospitable surroundings? This might well be a difficulty but for the fact that the early forms of a seedling plant may be maintained even without specially vigorous growth by giving the plant a certain environment

[Footnote] * It was Mr. R. Brown (Brown ter.). who first demonstrated the true character of the so-called leaves of that section of Veronica, in a paper read before the Horticultural Society of Christchurch, but not published in any recognised scientific publication.

– 358 –

which possibly may represent in Some degree that of the hypothetical ancestor. Thus a seedling plant of Veronica armstrongii, Kirk, pot-bound, and so not subject to vigorous growth, but kept in the mild and somewhat moist climate of my greenhouse, still, although more than two years old and 12cm. high, keeps in every part its seedling form, and may very well indeed be a, veritable reconstructed ancient type (see Plates XXVIII. arid XXIX.). The case cited above of Veronica tetrasticha may be explained in a similar manner; and finally the leaves may also serve as organs for promoting rapid growth without upsetting the theory of reversion, since the former ancestor, if provided with such leaves, and living under more favourable conditions, would certainly grow much faster than its depauperated descendant. Also, quite a number of plants never quite lose, or, at any rate, keep for a very long time, these seedling forms, which exist side by side under the same conditions with the mature, sometimes even on the same shoot. Parsonsia (all the species), Dacrydium (all the species), Muhlenbeckia complexa and M. adpressa, Gaya lyallii, and Hoheria angustifolia afford good examples, while Pseudopanax is more remarkable still, since it goes through at least three different forms, and does not assume the last form—most distinct from the preceding, but which is not much unlike the early leaves—until quite late in life.

That such great instability of species should occur in our flora is not to be wondered at when we consider—with Captain Hutton and those later writers who have followed his theories—the diverse origin of our flora, and the comparatively recent great changes which have taken place since all our plant immigrants finally settled down in their island home. This insular habitat, as pointed out by Diels, would, when a change in climate came, cause the survival afterwards of various ancient forms in stations not suitable for their well-being,* whereas in a continent they would under similar circumstances either migrate to a suitable spot or be wiped out by forms better suited to the new conditions. In such an unfavourable locality these survivors would, I may venture to suggest, be modified more or less in outward appearance, but would never become really stable species. A good case in point is Raoulia bryoides and Raoulia mammillaris, both of which, under cultivation in a greenhouse, develope true leaves, and become in habit much more like their relatives of

[Footnote] * Carmichælia crassicaule, Raoulia mammillaris, R. eximia, R. bryoides, Haastia pulviaris, H. recurva, H. sinclairii, most inhabitants of shingle-slips, Veronica (of the whip-cord and epacridea type), V. buxifolia, Ozothamnus, Panax lineare, Notospartium, Epilobium crassum, Hymenanthera alpina, Carmichælia monroi, C. robusta (?), Senecio cas-sinioides, & c., are probably examples of such forms.

– 359 –

the river-beds and plains. These latter, indeed, to judge by the seedling leaves of the former, should be the ancestral forms, but it is quite likely that they are reproduced ancestral forms from these now almost defunct species. Such might be called reinstatement of a species, and the Pseudopanax quoted before may also be explained on this supposition.

The unsuitability of many well-known and common plants to their surroundings seems worthy of mention here. That very little change in climate would be required to destroy for ever some of our indigenous plants may be seen by their, behaviour in cultivation. On the low-lying portions of the Canterbury Plains quite a number of South Island plants are not really hardy. Fuchsia excorticata, Aristotelia racemosa, Piper excelsum, Schefflera digitata, and Myoporym lœtum will at once suggest themselves to any grower of New Zealand plants. Others—Veronica elliptica, for instance—are easily damaged by frost. From my paper, “On the Freezing of Alpine Plants” (Transactions, vol. xxx., p. 435), it will be seen that quite a small amount of cold, and that for no very long period, was sufficient to kill fifteen out of twenty-two, and the survivors were nearly all so much damaged that they could hardly hope to survive. A severe winter in a climate so mild as England plays great havoc with most New Zealand plants, many Veronicas being hardy only in such localities as Cornwall, Devon, the Isle of Wight, &c. At Castle Hill, in a sheltered ravine, after a severe winter, I. have noted Veronica traversii, var., very much damaged, and at 1,460 m., on Hill's Peak, on a ridge fully exposed to the wind, at one time covered with vegetation, all was destroyed during the winter of 1896.

It may not be out of place here to give some account of how I have carried on, and propose to carry on, this work. My pursuits as field-naturalist and collector, gathering yearly as many different sorts of seeds as I can procure, puts me in possession each season of seeds of from one hundred to two hundred species of plants, the sources whence all of them were procured being exactly known. These seeds I sow in small pots in very porous sandy loam, keeping those species well apart which belong to the same genera, so that no confusion can arise through seeds being washed from one pot into another. Small seeds, such as Gaultheria, Fuchsia, Pratia, and even larger ones, such as Veronica, Celmisia, and other Compositæ, I do not cover with earth, for they germinate much better when firmly pressed down upon the surface soil and kept moist. Each pot is labelled with name of plant, date of sowing, and a number which corresponds to the one on the seed-packets as sent to various botanic institutes

– 360 –

with which I correspond; and such number is entered—with an account of the environment of the plant from which the seed was gathered—in a book kept for that purpose. The pots are plunged in sand on the staging of an unheated greenhouse, at no great distance from the glass. The germination of each species is recorded as it occurs. So soon as the cotyledons are developed a magnified drawing is made of them, and a description written; and further descriptions are written and drawings, magnified or not as the case may be, together with nature prints and photographs, are made of the succeeding leaves as they develope. This is kept up until the plant assumes its final form, when it is planted in the garden, or potted and kept in a shade-house for reference, & c. Of course, these proceedings take up a good deal of time, so the work is slow, very few of the New. Zealand plants being of quick growth; the seeds, too, are often slow to germinate. So, without waiting for the full results in each particular case, I propose to publish from time to time, in future parts, my results up to date. Nor can there be any systematic arrangement, since plants of most varied natural orders are ready for description at the same time.

Of course, seedlings grown under such artificial conditions may not be identical with those which occur spontaneously in their natural habitats, and I am in the habit of collecting, describing, potting, and noting the changes of the latter, if any, or the differences from the artificial seedling. Dr. Dendy has suggested to me that the first stage of an artificial seedling's existence should not differ in form from one growing naturally, since the embryo in the collected seed was developed under its natural environment, but that a seedling raised from cultivated seed might well show some difference. This suggestion is most interesting, and the differences between two such seedlings should be carefully noted.

The environment of the wild seedling will, when possible, be described, since this differs in much from that of its parent; and although, as I pointed out above, most seedling forms are in all probability, ancestral, yet at the same time they may also be adaptations for the benefit of the young plant. Abundance or the contrary of the seedlings of any species, the struggle for existence between the same or different species, and other matters which bear on my subject will be dealt with in due course.

I cannot conclude this paper without referring to my friend, our member, Mr. R. Brown. Some years ago he read a paper on the leaves of the whip-cord Veronicas before the Horticultural Society, which was, however, somewhat incorrectly published. His views, which in many respects differ widely from mine, have influenced my thoughts in no small

– 361 –

degree, and I must express how much indebted I am for his early suggestions. Amongst other things, he raised plants of Veronica armstrongii from seed, caused Ozothamnus microphyllus to revert to what I should call its ancestral form, and grew some very curious forms of Raoulia, which are, unhappily, lost. To Captain F. W. Button, for assistance regarding literature relating to geological history of our flora, to Professor A. Dendy, for the use of seedling plants, and to Mr. W. Sparkes, for photographing V. armstrongii for the accompanying plates, I must also beg to record my thanks.

Part II.: Descriptions of Seedings, and Notes Thereon.

With regard to the seedlings described here, it may be well to point out that, where nothing is stated to the contrary, they have been raised by myself in the manner and under the conditions described in Part I. of this memoir, and so it cannot be held that such seedling forms are identical with those to be found wild. Such wild forms I hope to describe or to remark on when opportunity offers; and the comparison between these and artificially raised forms, showing the effect of various environments, should be of interest. As for the measurements given below, since the lengths of most parts of a seedling vary at different periods of its development, such measurements must be looked on as comparative, and serve only to give some idea of the relations in point of size between the different parts of a seedling at some fixed period in its development, and are not in any case a uniform measure for any particular part. Measurements of leaves always include the petiole unless the contrary be stated. With regard to the greater part of the forms described, their development up to the present is not sufficiently advanced to warrant many remarks; the further development of such will be of much greater interest; an account of this I hope to publish in Part III. As for the descriptions, they are drawn up from an examination of all the seedlings raised as far as possible, and are not descriptions of one seedling alone. Many seedlings die after having been examined, while others must of necessity be cut up for examination and so destroyed. Thus it has not been found possible to watch the development in many cases of any particular individual, the further progress of the species having to be noted from the survivors. Besides using herbarium specimens, I have been enabled to compare nearly all the seedling forms with adult plants from my garden. As this garden is of dry sandy soil, considerably exposed to wind

– 362 –

and sun, the majority of plants from the drier regions of New Zealand are growing under much the same conditions, altitude excepted, as the wild plants, and their organs cannot be very different in appearance.

No. 360. Pittosporum rigidum, Hook. f. Plate XXX., fig. 4.

Seed collected at Craigieburn, at altitude of.602 m., from a shrub growing on a very dry stony slope, fully exposed to high winds and sun. Sown 12th December, 1897; germinated early in the spring of 1898.

Description of Seedling.

(Development not yet nearly concluded; most mature plant 4.5 cm. high, and with eight leaves.)

Root long, straight, deeply descending; side rootlets few, short.

Hypocotyle 1.8 cm. or more long, straight, bent or twisted towards base, terete, reddish, hairy with brownish short glandular hairs.

Cotyledons three in a whorl, almost sessile, articulate at base; pulvinus dark-purple; lamina 8 mm. × 3 mm., linear-oblong to linear-lanceolate, with sides sometimes unsymmetrical, obtuse or acute, pale yellowish-green; upper surface concave where it merges into the short channelled petiole; midrib evident above and beneath; veins of upper surface often swollen; margin entire, sometimes irregularly waved, ciliated.

Leaves cauline, alternate, cuneate at base, deeply toothed or pinnatifid, usually acute, almost glabrous except on the slightly hairy petiole, or with one or two scattered hairs near the base of leaf; most recently developed leaves crowded towards apex of stem.

1st leaf rarely nearly opposite 2nd leaf, ovate or obovate, 8 mm. × 5 mm., thick, rather coriaceous, green, polished, pinnatifid or very deeply and coarsely toothed with blunt teeth; segments unevenly serrate.

2nd to 5th leaf similar to 1st leaf, but varying very considerably in breadth and length; largest 1.3 cm. × 5 5 mm.; petiole 3 mm.; segments often subulate; later leaves often conspicuously edged with reddish-brown, stained dark at base, and of much duller darker green than earlier; laminæ usually patent and horizontal, but sometimes semi-vertical.

Stem terete, very dark-purple in oldest portions, lighter purple above, covered rather thickly with coarse adpressed white hairs; 1st internode 2 mm. long, remainder varying in length from 2 mm. to 4 mm.

– 363 –

The seedlings examined were remarkably uniform, varying very little from one another. The leaves of adult plants are smaller, dark in colour, much less divided or quite entire, more coriaceous, with obtuse apices and recurved margins, often fascicled on very short branches, and are evidently reduced forms of the greener, larger, pinnatifid, acute, juvenile leaves. This reduction seems, to afford a striking example of the effect of environment on leaf-forms—in this case frequent furious gales, no shelter from frost, often insufficient moisture, a very clear atmosphere, fierce sunshine, and very poor stony ground. And it is worthy of remark that the majority of the shrubs growing in company with P. rigidum have sometimes reduced leaves almost identical in shape, and at other times leaves of different shape but still much reduced. Such, for example, are Coprosma (several species), Hymenanthera (several species), Panax anomalum, Veronica (several species), Discaria toumaiou, Clematis marata, Rubus cissoides, and Aristotelia fruticosa, this last having in one of its early forms pinnatifid leaves.

A form of P. rigidum from Mount Hikurangi, kindly sent to me by Mr. D. Petrie, P.L.S., has much larger leaves, obovate or oblong, and resembles the plant figured in the “Flora Novæ Zelandiæ” (vol. i, pl. x.) much more than the southern form. It is quite possible that this Canterbury form, here treated of, may be specifically distinct from that of Nelson and the North Island, but, until seedling forms of these latter are studied and experimented with, it will be impossible to settle the point. Seedlings of the little-known P. obcordatum would also be of great interest, and would help to throw light on these anomalous forms of Pittosporum.

No. 504. Carmichælia hookeri,* T. Kirk. Plate XXX., figs. 6, 7, 8.

Seed gathered by Mr. T. Kirk, F.L.S., at head of Porirua Harbour. Germinated in about three weeks.

Description of Seedling.

Root stout, with many fibrous strong lateral rootlets.

Hypocotyle very short, thick, green, terete.

Cotyledons increase in size after germination, from 4 mm. to 7 mm. in length, obovate or oblong, from dark to pale green, paler on under-surface, thick, fleshy, short-petioled; petioles finally 1 mm. to 2 mm. long, channelled, connate at base, slightly hairy or glabrous; sides of lamina unsymmetrical; midrib just evident on under-surface, or slightly swollen towards base of leaf.

[Footnote] *“On Carmiohælia,” &c., by T. Kirk, F.L.S. (“Transactions of the New Zealand Institute,” vol. xxix., pp. 506, 508).

– 364 –

Leaves (Plate XXX., fig. 7) distichous on edge of stem or cladode, becoming much depauperated as development proceeds, cauline, petiolate, stipulate, at first green, then variegated with light stains at base, finally brownish-purple.

1st leaf (Plate XXX., fig. 8) rotund, subrotund or obcordate, variable in size; lamina from 1 cm. × 1 cm. to 7 mm. × 6 mm., rounded or slightly cuneate at base, emarginate, ciliate with a few scattered deciduous (?) hairs, pale-green on upper but much paler on under surface, very sparingly pilose: with a few adpressed white hairs; margin entire, often purplish; midrib reddish at base; veins evident below; petiole 7 mm. long, channelled above, terete beneath, articulated at junction with midrib, the articulation becoming more evident as development proceeds, slightly hairy, articulated at base, and furnished with two short membranous, triangular, adnate stipules, which are purple and hairy, especially at apex.

2nd leaf similar to 1st leaf, but usually rather smaller.

3rd leaf as 1st or 2nd leaf, or sometimes as 4th leaf.

4th leaf ternate with three obcordate leaflets, similar in hairiness, &c., to the previously described leaves; lateral leaflets 2.5 mm. × 2mm., often larger, articulated to midrib; terminal leaflet much larger than lateral, 6 mm. × 4.5 mm.; petiole 6 mm. long, similar to those described above.

Stem usually at first grooved and ribbed, sometimes 4-ribbed or merely angled almost glabrous, soon becoming through flattening of its surface and reduction of leaves a leafy cladode.

Leafy cladode (Plate XXX., fig. 6) dorsiventral in all the seedlings raised except one;* upper surface terete, 4mm. wide, furrowed with usually six to nine shallow furrows, dark-purple, dotted with minute white scales, canescent with short white bristly adpressed hairs in adjacent patches; ridges paler than furrows, translucent; under - surf ace flat, much less hairy than upper surface, or almost glabrous, except for dotting of minute scales, never canescent, always much grooved; margins pink, translucent, much notched; notches rounded, with pale-coloured upper margin swollen and extending as a transverse striation or swelling half across the stem on both surfaces; internodes, in plant 13.5 cm. long, 6 mm. long, 3 mm. broad at base, 4 mm. broad at notch, 2 mm. thick.

Further development: The succeeding leaves are more and more depauperated, the lateral leaflets being smaller

[Footnote] *This exceptional seedling differs chiefly in the upper surface not canescent, and in its much thinner texture; its under-surface, however, is more distinctly furrowed than the upper.

– 365 –

and smaller, until one, and then both, are suppressed, only a very small terminal leaflet being produced. Finally leafless cladodes are developed from the lower nodes, sometimes at first furnished with one or two minute leaflets.

Leafless cladodes flat, dorsiventral, distichous, usually curving downwards, notched at nodes, furrowed on both surfaces but more so on under-surf ace, canescent on upper surface, often stipulate on upper margin of notch.

The leafless species of Carmichœlia seem to exhibit usually three stages of development—first, a wiry (Plate XXX., fig. 9) or thick - stemmed (Plate XXX., fig, 8) seedling, with simple or compound leaves; second, leafy cladodes arising from the axils of the primary leaves; and third, quite leafless cladodes, often extremely thick and stout. The second form may be artificially produced in an adult leafless plant by cultivation in shady sheltered, moist situations (see Plate XXXI, fig. 13); such also occurs spontaneously in a state of nature, but I have not had an opportunity as yet of accurately investigating such a change. Certainly shade and its accompanying moisture is here also a factor. C. flagelliformis, on the Port Hills, is in full sunshine leafless or almost so, and in shade of trees a quite leafy plant. Compare also the permanently leafy (deciduous in winter) C. grandiflora, growing only in positions where it receives the full western rainfall, it being the common Carmichœlia of Westland, with the leafless C. robusta* of the Trelissick Basin.

Carmichælia gracilis, Armstg. Plate XXX., figs. 1, 2. (Syn. C. kirkii, Hook. f.)

Seed collected from one plant, growing in rather swampy ground, at foot of sand-dunes near New Brighton, Canterbury. Germinated in about ten days.

Description of Seedling.

Root long, straight, white, fleshy at base, with numerous lateral rootlets.

Hypocotyle stout, very rigid, terete, tapering from the base. upwards, smooth, glabrous, pale, tinged at times with purple, 5 mm. long, partly subterranean, at times slightly flexuous.

Cotyledons extremely fleshy, persistent for a considerable time, increasing in size, at first 7 mm. to 8 mm. × 4 mm. to 5 mm., finally even 1–6 cm. × 9 mm., and 1.25 mm. thick, un-symmetrically oblong or obovate-oblong, entire, with purple

[Footnote] * Of course, it must be a matter of doubt what plant was meant by Kirk as C. robusta without access to his unpublished work, and to his herbarium, lately acquired by the New Zealand Government.

[Footnote] † “Descriptions of New and Rare New Zealand Plants,” by J. B. Armstrong (Trans. N.Z. Inst., vol. xiii., p. 336).

– 366 –

margin, glabrous or with a few scattered hairs on under-surface; upper surface flat; under - surface convex; petioles short, slightly channelled above, connate at base.

Early development: While the first leaves are in the bud, or slowly unfolding, the stem is being rapidly developed, attaining in some specimens a length of 2 mm. By the time the plant has attained a height of 2.45 cm. it possesses three leaves fully or nearly fully developed, from the axil of each of which, and also from the axils of the cotyledons, a shoot is being pushed forth.

Leaves cauline, distichous, large, inserted on margin of stem when latter is flattened, long-petioled, stipulate, spreading, with upper surface of leaflets horizontal.

1st leaf (Plate XXX., fig. 2) 1.3 cm. long; lamina 5 mm. or 6 mm. × 4 mm. to 6 mm., obcordate or rotund-emarginate, entire with red margin, yellowish-green but rather paler on under - surface, glabrous; nerves not visible; midrib raised considerably on under-surface; petiole articulated to midrib at about 5mm. from base of lamina, semi-terete, channelled above, brownish, articulate at base.

2nd leaf variable in size, in some cases 2 cm. long, ternate, each leaflet articulated to petiole by a prolongation of its midrib; leaflets very similar in all respects to 1st leaf; petiole articulated at base, furnished with a few scattered hairs.

Succeeding leaves (Plate XXX., fig. 1)—largest plant examined 28 cm. long and with fourteen leaves—very similar to 2nd leaf, but lateral leaflets usually smaller than terminal, arid rarely very much reduced, variable in colour from bright-green with brown margin to brownish dotted minutely with pink and pink margin and midrib; laminæ almost glabrous, with a very few hairs on the margin or midrib; petioles with scattered white bristly hairs on margin.

Seedling leaves as compared with adult leaves: Seedling leaves are much larger, thinner, and paler green than adult leaves, which, moreover, become smaller and smaller towards tip of branch.

Stem dark-purple, semi-erect or decumbent, at first wiry, terete, almost glabrous or marked with minute white scales in parallel longitudinal rows separated by hardly raised glabrous ridges. As development proceeds from the 3rd inter-node upwards, or even sooner, the upper surface of the stem becomes quite hoary with adpressed white bristles, and at the same time is slightly flattened and broadened, but not flat; under-surface marked with ridges and lines of scales more evident than in early internodes, slightly hairy; internodes very constant in size, lengthening with development. Of plant 28 cm. long they are 2 cm. long and 2 mm. broad; of plant 10 cm. long they are 1.6 cm. long.

– 367 –

Stipules triangular, lacerate or bristly at apex, adnate, embracing base of petiole.

In its natural habitat C. gracilis is a plant with long straggly usually quite leafless stems, spreading and becoming entangled in the neighbouring vegetation and itself, after the manner of some of the New Zealand forms of Rubus and Muhlenbeckia, which in habit it much resembles, such being very different indeed from that of typical species of Carmichœlia. Where the stems are sheltered adult leaves are developed as described above. Cultivated in a moist greenhouse a plant rapidly developed shoots not unlike the juvenile form, but when this same plant was removed to a very dry situation in my garden the new stems that were produced were almost, leafless. The seedling form puts one in mind of C. exsul, of Lord Howe Island (vide fig. 4A, page 266, in Diels's work*) and it looks far more adapted for a moist forest region than for a lowland swamp liable to drought in summer.

Carmichælia crassicaule, Hook. f. Plate XXXI., figs. 12 and 13 (Corallospartium, Armstg.).

Seed collected at Mount Torlesse, from plants growing on stony hillside below the winter snow-line, in dry clayey soil, and exposed to wind and sun. Germinated in about four weeks. The seed was quite soft when gathered.

Description of Seedlings.

(The most developed of the seedlings were grown by Professor A. Dendy, D.Sc., and kindly given me for this work.)

Root long, deeply descending; lateral rootlets few, short.

Cotyledons 6 mm. × 4 mm., unsymmetrically oblong, sometimes falcate, obtuse, fleshy, entire, with faint-red margin, glabrous, darker green on upper than on under surface; petioles connate at their base, semi-terete, channelled above, pointing upwards and outwards; lamina with surface quite horizontal.

Stem notched at nodes; 1st internode variable in length, 3.5 mm. to 5.5 mm., almost terete at base, slightly angled, dotted with minute scales in young plant 1.9 cm. high, and hairy with long matted white hairs in plant 4.3 cm. high; 2nd internode in the younger plant almost glabrous and more deeply grooved than 1st internode, in the older plant similar to 1st internode; 3rd internode hairy, as described for older plant. In the older plant the stem is much thicker than in the younger, purplish-brown, with very deep furrows and prominent longitudinal ridges marked with one or two longi-

[Footnote] *“Vegetations-Biologie von Neu-Seeland,” L. Diels, Leipzig, 1896.

– 368 –

tudinal purple lines; margin furnished with a narrow pink transparent wing on each side.

Leaves cauline, entire, narrow-ovate, broad-ovate or almost rotund, at first opposite, afterwards distichous through flattening of stem, shortly petioled, dull dark rather glaucous green on upper surface, much paler glaucous on under-surface, stipulate with two membranous broadly triangular hairy stipules; margin edged with reddish-brown; midrib sunken on upper keeled on under surface; veins sometimes swollen on upper surface and much reticulating; apex retuse or emarginate; petiole jointed at junction of lamina and midrib.

1st leaf 6 mm. × 4 mm., in one plant much larger and rotund-emarginate.

2nd leaf sometimes rather smaller than 1st leaf.

3rd leaf a little larger than 1st leaf.

4th leaf 1 cm. × 8 mm.

Later developed leaves more hairy than those preceding.

Further development: In the largest seedling examined, 5 cm. high, flattened hoary cladodes are being developed, longest 2 cm. long, with four leaves similar to but smaller than those on the main stem; laminæ semi-horizontal.

Leafy cladodes on adult plant (Plate XXXI., fig. 13). Under the influence of shade, moisture, and shelter, also perhaps of heat, leafy cladodes are developed with extraordinary rapidity, growing as much as 8 cm. in a space of four weeks, with internodes 2.5 cm. long, 5 mm. broad, 3 mm. thick, shallowly grooved; grooves filled with white, slender, weak, straggly hairs; ridges glabrous; nodes much notched, swollen, and rounded, from each of which arises a small, narrow, linear-oblong, emarginate, fleshy leaf, cuneate at the base, with sides inflexed and surface reflexed from centre. Similar cladodes are developed, sparingly on plants in their natural habitats during early spring.

Leafless cladode on adult plant deeply grooved; grooves separated by longitudinal ridges 1.5 mm. wide, yellow towards margin and marked down centre by a green line; grooves filled with dense white matted tomentum; internodes 1.5 cm. long × 9 mm. broad × 6 mm. thick; nodes marked by deep notches furnished with two depauperated stipules, or O. Such a cladode was developed in my garden on a plant in the open air, partly exposed to north-west and south-west wind and full sunshine, while its sister plants—one in a shade-house and the other under a bell-glass, and kept very moist—developed leafy cladodes as described above.

The development of C. crassicaule is of great interest. Apart from its being one of the most singular plants of which our rather anomalous flora can boast, it is, although widely

– 369 –

spread, extremely local and limited in its distribution, and is confined to most inhospitable situations on the mountains at an altitude of from 690 m. to 1,200 m. Wild seedlings are unknown to science, and, although such doubtless exist, their quantity roust be small, and young plants also are very scarce. The plant, however, adapts itself very readily to cultivation, and, as stated above, quickly responds to change of environment by developing leafy cladodes, which must assist its growth very considerably. It seems possible from a consideration of these facts that the wild plant was once much more common, and that it is now dying out. Also, that if not an ancient type in the sense that it has existed from a remote geological time, still it may very well be the survivor of a race whose prime dates back for a very considerable period, and whose days are now numbered. (For microscopic drawing of stem, see Diels, page 266, fig. 4c.) The hairs simply covering the stem, often somewhat sparsely in the young plant, and afterwards being confined to the furrows in which lie the stomata, is very interesting.

Carmichælia odorata, Col. Plate XXXII., fig. 24.

Seed-from garden, of Dr. E. G. Levinge, Sunnyside; the plant said to have been originally brought from the Hawke's Bay District. Germinated in about four weeks. Tallest plant examined, 2.5 cm. long, and with three leaves.

Description of Seedling.

Root white, straight, deeply descending, with few lateral rootlets.

Hypocotyle mostly subterranean, 7mm. long; subterranean portion fleshy and white, thicker than the green short aerial portion.

Cotyledons oblong, obtuse, fleshy, dotted with numerous white scales, light-green, tapering into the very short broad connate petioles.

Stem terete, grooved and ridged; ridges narrow, glabrous, translucent; grooves even, dotted with scales similar to those on cotyledons; internodes flexuose, in tallest seedling 1st is 5 mm. and 2nd is 4 mm. long.

Leaves distichous, bright light - green, almost glabrous, except for scales as on cotyledons, stipulate with triangular sometimes lacerate hairy stipules; petioles long, semi-terete, channelled above, articulated at base.

1st leaf: Lamina 7.5mm. × 6 mm., entire, oval or sub-rotund, emarginate, green above, much paler beneath; midrib slightly sunken on upper raised on under surface; petiole 6 mm. long.

2nd leaf ternate (10 mm. long) or imparipinnate (1.3 cm.

– 370 –

long); leaflets entire, obcordate, glabrous, except for a few scales, jointed to petiole, cuneate at base; apex emarginate or retuse, with sometimes a slight mucro in the centre of the sinus.

3rd leaf similar to those described, but not yet fully developed at time of writing.

As the development of the seedlings under examination is not yet far enough advanced, I have reserved comparison with mature form for future occasion.

No. 514 Ozothamnus depressus, Hook. f. Plate XXXIV., figs. 50 and 51.

Seed collected from plant growing on stony bed of River Kowai. Germinated in about four weeks. The seed germinates very freely, but it is difficult to keep the young plants alive for any lengthy period.

Description of Seedling.

Hypocotyle at first succulent, very early on becomes woody, and is already 4 mm. long, as the tip of the first leaf can just be seen arising from between the cotyledons.

Cotyledons (Plate XXXIV., fig. 51) 2 mm. long, broadly oblong, obovate, almost obcordate at times, sometimes unequal in size, emarginate, obtuse, retuse or truncate, indistinctly pubescent with very short white hairs, sessile, amplexicaul, at first erect protecting the 1st leaf, then becoming more patent.

1st pair of leaves (Plate XXXIV., fig. 50) 2.5 mm. long, linear-oblong to linear-spathulate, amplexicaul, very succulent, densely woolly with long matted hairs which are erect and adpressed to stem below, patent above, with. apex curving downwards.

2nd pair of leaves.3 mm. long, similar in other respects to the 1st pair.

Stem terete, very soft and succulent, very pale-green, woolly but not so much as the leaves; 1st internode 1.5 mm. long; 2nd internode 2 mm.; or the internodes may be almost suppressed or very indistinctly seen, owing to the leaves being so closely adpressed to the stem. Before the opposite leaves open out their hairs are all matted together into one mass.

In the mature plant the leaves are most closely adpressed with ventral surface to stem, and covered, as in seedling, with dense matted cotton; the dorsal surface, also equally tomentose in the seedling, is in the adult covered with a silky pellicle, and the internodes are almost suppressed. Plants cultivated in a shade-house have put forth shoots with adpressed leaves, but green on the dorsal surface, much less

– 371 –

hairy than the seedling leaves, but in other respects almost identical; so far, however, I have not succeeded in getting a shoot to revert to the actual seedling condition. O. depressus is one of the most extraordinary-looking plants in New Zealand; even a most healthy plant, with its stunted, rigid, naked branches and small scale-like hoary adpressed leaves, seems to the eye hardly to possess life. It grows on dry river-beds in subalpine or even in quite low regions, ascending sometimes, but rarely to the alpine zone, where it occupies the most arid and barren regions, forming in the former situation considerable colonies. I have never observed wild seedlings, but that must be more from my not having known the seedling form than from their scarcity. The plate in the “Flora Novæ Zelandiæ” gives very little idea of the habit or appearance, and represents an altogether more natural looking plant. I have never seen plants of anything like the size—5ft.—of that mentioned in the “Handbook of the New Zealand Flora” (page 146).

No. 435. Rubus australis, Forst., var. glaber, Hook. f. Plate XXXII., fig. 21, and Plate XXXIV., figs. 67 and 68.

Seed collected from plant growing at the base of Mount Earnslaw, Lake Wakatipu, at altitude of 340 m.

Description of Seedling.

Root fibrous, descending, sparingly branched.

Hypocotyle 4 mm. or 5 mm. long, pale, terete, fleshy, glabrous or very slightly hairy.

Cotyledons narrow-oblong, 4 mm. × 2 mm., acute with slightly swollen tip, ciliated with transparent white glandular hairs; petiole very short, erect, with lamina at right angles.

Stem terete, reddish; internodes at first very short, lengthening as development proceeds, furnished with very slender white or brownish hairs and numerous glands with red swollen tips (Plate XXXIV., figs. 67 and 68).

Leaves alternate, extremely membranous.

1st leaf ovate-cuneate, rotund or broadly oblong in outline, 8 mm. × 4.5 mm.; margins wavy or entire on lower and deeply toothed on upper half, with glandular swollen red teeth; lamina sparsely hairy with glandular hairs, as on stem; venation distinct, reticulated; petiole subterete, channelled above, three-fifths length of lamina.

2nd leaf very similar to first, but with venation more marked.

3rd leaf ovate, deeply toothed except at rounded base, almost glabrous except on under-surface; midrib and nerves swollen below, purplish, especially midrib; petiole channelled, sheathing at base.

– 372 –

Succeeding leaves very similar, but become more triangular, often ovate-lanceolate, more rugose above, while below the veins form a complete pink network, sometimes with reddish-brown round apex, and sometimes cordate or subcordate at base; each young leaf as it developes at first perfectly bristles with hairs.

The extreme thinness of the seedling leaves as compared with those of the adult plant is of interest. The seedlings have been grown in a moist greenhouse, and abundantly supplied with water. A prostrate form of Rubus recently brought by Mr. S. D. Barker from Glen Bonnie, Westland, and growing in the forest in complete shade, has mature leaves remarkably like those described above, while in the very wet forests of that district the common Rubus (possibly R. cordata,* Armstg.) has leaves of a very thin texture. The adult form of the seedling plant described has leaves coriaceous and stiff, with much smaller toothing (in comparison to size of leaf), very smooth and polished on the upper surface, a leaf-structure well suited to resist the drying influence of almost constant winds.

Sophora microphylla, Ait. Plate XXXII., fig. 25.

Seed gathered from plant growing on bank of Broken River, near its junction with the Waimakariri. Germinated very irregularly, some in four weeks, some in twelve weeks, and some took much longer still.

Description of Seedling.

Cotyledons hypogeal, thick, fleshy, obovate, obtuse, falcate, plano-convex, with the flat sides together and not separating, 9 mm. × 6 mm., and the two together 7 mm. thick.

Stem at-first tapering from the base, subterranean part paler white than aerial, grows very rapidly at first, 2.3 cm. long before any leaves are fully developed, afterwards yellowish-brown, scaly and hairy with adpressed hairs, at first straight, then flexuose, branching at each node at first; young twigs extremely bristly, with adpressed greyish-white bristles; internodes, in plant 9cm. high, at first 1.1 cm., later ones longer, 2.1 cm.

Leaves: First few simple; afterwards ternate and imparipinnate, exstipulate, petiolate, bright yellowish-green, alternate; leaflets obcordate, varying in width and size, articulated to petiole, slightly hairy on under-surface, penninerved; nerves and midrib raised beneath; under-surface pale bluish-green; margins faintly incurved, thickened; petioles channelled, hairy, semi-terete.

[Footnote] * Trans. N. Z. Inst., vol. xii., p. 338. Merely the name is given, with no description except “Leaves cordate.

– 373 –

1st leaf broadly linear, curving-downwards, sides incurved, 5mm. long.

2nd leaf usually simple, but larger or sometimes ternate, with, one of the lateral leaflets much reduced.

5th leaf, two lateral and one terminal leaflet, which as rather larger than the lateral; leaf 1.6 cm. long; lateral leaflets 6.5 mm. × 5 mm.

This species was included by Hooker with S. tetraptera, Ait., as var. β microphylla, although he distinctly states that the two forms are very distinct, and remain so under cultivation.* Kirk also considered that there was only one species of Sophora in New Zealand, and divided it into three varieties and one sub-variety. The seedling form, however, of the plant under consideration differs much from that of S. tetraptera, var. grandiflora, its stem being more slender and flexuous, branching at an earlier age, in colour more yellow, and with, leaves and leaflets much smaller—e.g., the 5th leaf in S. microphylla is 1.6 cm. and lateral leaflet 6.5 mm., while in S. grandiflora (Plate XXXIV., fig. 71) it is 2.5 cm. and the lateral leaflets 1.2 cm. A form from the Chatham Islands, given to me by Mr. S. D. Barker, seems also distinct in its seedling form (I have only one plant). This early on developed more leaflets in each leaf than in either of the above, while in size these are intermediate between the two (Plate XXXIV., fig. 70). Seeing these well-marked differences in the juvenile forms of Sophora in New Zealand, I hold that there are certainly two, and perhaps four (S. prostrata), quite distinct species, which will possibly vary very considerably; nor is it unlikely that they may hybridize, and so produce some of the so-called “intermediate” forms.

No. 341. Veronica cataractæ, Forst.

Seed from cultivated plant in garden of Mr. A. Bathgate, Dunedin; the plant originally from one of the West Coast sounds.

Description of Seedling.

Hypocotyle short, pale, terete, glabrous, fleshy, 2–3 mm. long.

Cotyledons rotund-oblong, obtuse or subacute, fleshy, rounded at base of lamina, glabrous, pale; petioles very short, connate at base.

1st leaf 2.7 mm. × 1.5 mm., oval, acute, entire, glabrous; petiole short.

2nd leaf similar.

[Footnote] * “Handbook of the New Zealand Flora,” p. 53.

[Footnote] † “Forest Flora,” p. 85.

– 374 –

3rd leaf: Shape similar, but with petiole nearly as long as lamina.

Up till about the 6th leaf the newly produced leaves are similar to the above, except that leaves with one or two teeth on each side make their appearance.

6th leaf 1.4 cm. × 8 mm., obovate, lower third cuneate and entire, upper two-thirds serrate, with four or more serrations, acute, pale shining green above, much paler beneath; midrib sunken; petiole one-third as long as the leaf, channelled, with two minute, adnate, hairy, chaffy stipules at base.

Stem terete, pale, with internodes variable in length, bifariously hairy, the two rows on each internode alternate with those on the succeeding; hairs very close, and curving upwards at the tip; adventitious roots are early formed, coming from the lower nodes.

Variations not many; the petioles may be shorter than those described for the 5th leaf, and the laminæ oval, and thicker in texture.

The seed germinated very freely, and the seedlings became much crowded. On this in large measure depended the variations, length of internodes, and texture of leaves. The plants grow very slowly indeed, much more slowly than the shrubby Veronicas, and this is surprising when we consider the moist misty atmosphere of their native habitat.

Veronica salicifolia, Forst. Plate XXX., figs. 3, 3a, 3b.

Seed gathered from plant on bank of Broken River by A. H. Cockayne, at altitude of 600 m. Germinated in about four weeks.

Description of Seedling.

Root at first straight, white, fleshy, soon becomes much branching from near junction with hypocotyle; branches wide-spreading, slender.

Hypocotyle white, fleshy, 3 mm. long or less, lengthening very little as plant grows, terete, minutely hairy.

Leaves vary somewhat, especially in early stages, bright-green, polished above, slightly toothed or quite entire; later leaves conspicuously and evenly serrated with serrations 1 mm. long; petioles of early leaves one-third length of lamina, or even more, in later leaves hardly one-seventh as long until finally the adult leaves are almost sessile; early produced leaves ovate, oval, or ovate-lanceolate, the succeeding leaves passing through a gradual transition of forms, and so becoming a little narrower until the final adult lanceolate leaves are alone produced.

1st to 6th pairs of leaves (Plate XXX., figs. 3a, 3b) are

– 375 –

of the early ovate type, minutely hairy, with one or two blunt teeth or sometimes conspicuously lobed at apex, tapering at base into the short, channelled, connate petioles.

7th pair of leaves and onwards are of the lanceolate type, varying a good deal in width in different seedlings.

Stem at first green and succulent, terete, pilose with hooked hairs with hooks pointing upwards; nodes swollen and purple; internodes, in plant 6 cm. high, very constant in length, about 6 mm. long; branches from base not very, numerous, and in young plants producing leaves of the early ovate toothed type.

The above description differs materially from that bf Lubbock,* but it is more than possible that the seedlings he described were raised from some form quite distinct from the above. V. salicifolia, Forst., is one of those species which Hooker held to include a great number of very different forms connected by supposed intermediates. Baron von Müeller went even further, and reduced all the New Zealand Veronicas to one species. Kirk, in his recent writings, has separated some new species from the Veronicas of this sectione.g., V. rotundata, V. latisepala, and V. squalida—and probably intended restoring some of those discarded by Hooker, for in the same article he refers to V. myrtifolia, Banks and Sol., and I think when the seedling forms of V. salicifolia, its branches and allies, are worked out several new species will have to be made. The adult form of leaf in the plant under consideration—a widely spread form in the Canterbury Alps—has the leaf much drawn out and narrowed towards the apex, being quite acuminate. The much greater toothing in the juvenile than in the adult form is of interest. It is suggestive that seedlings grown in very damp localities, such as in the subalpine scrub at the head of Otira Gorge, are toothed to ah extraordinary degree, and self-sown seedlings of a closely related form growing in my garden amongst ferns are also much toothed. The channelled connate petioles guide water falling on the leaves to the leaf-bases, where it lodges for a considerable time.

No. 388. Veronica traversii, Hook. f. Plate XXXII., figs. 27 and 28, and Plate XXXIV., figs. 62 and 66.

(Probably the typical form.)

Seed collected from a plant at Ribbonwood Creek, Craigieburn Mountains, at elevation of 700 m. Germinated rapidly.

[Footnote] * “On Seedlings,” p. 326.

[Footnote] † “The Vegetation of the Chatham Islands,” pp. 45–47.

[Footnote] ‡ Trans. N.Z. Inst., vol. xxviii., pp. 528–530: “On Veronica,” by T. Kirk.

– 376 –

Description of Seedling.

Root fibrous, branching very much from base, spreading.

Hypocotyle usually upright, sometimes slightly bent, succulent, pilose, especially towards axil of cotyledons, 9 mm. long.

Cotyledons broadly rotund-ovate, rarely oval, glabrous or with a few short hairs, obtuse or very rarely emarginate; petioles connate at base and thickened, forming a purple ring round the stem, subterete, slightly channelled aboye.

1st pair of leaves ovate to broadly ovate. Lamina 5.5 mm. × 4mm., green on upper reddish-brown to green on under surface, flat, hairy; margin ciliated, with hooked hairs pointing to apex of leaf; petiole half length of lamina, sub-terete, channelled above, connate at base, forming raised ring round stem, as in cotyledons.

2nd pair of leaves similar to 1st pair, sometimes with longer petioles.

3rd and 4th pairs from ovate to ovate-oblong, entire or with one or two teeth on each side towards the apex, gradually narrowed on one side at the base into the slightly channelled petiole, which is about two-thirds length of lamina; pale-green or subglaucous above, often deep-purple beneath, except along margin of leaf, which remains green. Hairs as in 1st leaf.

15th pair of leaves: At about this point the leaves begin to vary considerably from those below, and from similar leaves in other seedlings, in one case being lanceolate, 2–1 cm. × 6 mm.; in another obovate, 1.3 cm. × 6 mm.; and in a third case harrow-ovate, 1cm. × 4 mm.; all almost glabrous, or at any rate much less hairy bhan the earlier leaves.

Stem terete, pilose with hairs similar to those on the leaf; internodes (in plant 8 cm. high) 4 mm. long.

The seed, from which the above-described seedlings were raised was all gathered from one plant. Although, as might be expected, a certain sameness runs through all the seedlings, most marked differences occur among individuals at all stages, in extent of toothing among earliest leaves and in shape among later leaves:—in other words, the individual does not nearly produce itself true from seed. But, regarding the species, I have examined seedlings raised from three other individuals which were distinct from No. 388 and from one another. In each of these a certain type may be seen occurring most frequently, and giving a character to each pot of seedlings; but here also there are many individual differences, and it may be expected that when the adult form is reached there will be several plants so distinct as to look like different species. Among a number of wild seedlings that I possess

– 377 –

even greater differences occur—-differences of shape, colour, size, and toothing. This great variation from seed throws a little light on the fact that such a great number of distinct wild forms of this species are to be met with, a species whose individuals do not reproduce themselves with any exactitude, while the specie's consequently reproduces itself still less exactly. The consequence is that, while certain forms maintain themselves intact, new variations must be of constant occurrence. I hope to return to this subject on some future occasion. When I have examined more material, and experimented with it under varying conditions, then it may be possible to fix limits of variation, and to adduce some reason for the occurrence of certain forms. Cuttings from adult plants put forth shoots with leaves of the seedling type. The hairs on the young leaves tend to retain moisture on the surface of the leaves for some time.

No. 336. Veronica elliptiea, Forst.

(Var. with very large leaves.)

Seed gathered at Bluff, from plant, exposed to sea-spray. Only two seedlings. Germination was very irregular.

Description of Seedling.

Hypocotyle very short, hardly to be distinguished from the root.

Cotyliedons quite succulent, convex on under-surface, oblong, obtuse, with short thick connate succulent petioles.

1st pair of leaves oblong, obtuse, entire, minutely hairy; lamina rather large and tapering into the petioles; petioles connate and swollen at the base.

2nd pair of leaves oblong or obovate-oblong, obtuse, with one or two faint, teeth on each side; petiole snort and more narrow than in 1st leaf, minutely hairy.

3rd pair of leaves obovate, shining, 1-toothed on each side; teeth, opposite, half-way towards rounded apex, the toothing making a rounded lobe towards upper end of leaf; lamina patent, channelled down middle on upper surface; midrib evident on under-surface.

4th pair of leaves (in process of development) ciliated towards tips, and with two teeth on each side.

Stem pink, terete, bifariously hairy; internodes 3 mm. long.

No. 402. Veronica tetrasticha, Hook. f. Plate XXXIV., figs. 55, 56, 57, 58.

Seed collected from plants growing on the Craigieburh Mountains, on shingle slips, at altitude of from 1,200 m. to 1,500 m. Germination was slow.

– 378 –

Description of Seedling.

Root white, descending, at first with few rootlets.

Hypocotyle 3 mm. long, succulent, terete, white tinged with pink, mostly above ground.

Cotyledons 2 mm. long, very succulent, pale-green or purple-stained; laminæ oblong; petioles fleshy, connate at base.

Leaves spathulate, connate at base of petioles and sheathing, or with 1st leaves perhaps sessile, pale-green above, purplish beneath, covered above and beneath at regular intervals with erect stout hooked white hairs, with hooks turned upwards towards apex of leaf, obtuse, entire, adpressed to stem at base, then patent or semi-patent, finally apical portion curving downwards.

3rd pair of leaves 5 mm. × 2 mm.

Development of leaves (from examination of the growing-point): In an early stage the leaves are quite linear and sessile, very succulent, not marked with purple, closely imbricated round the growing-point, flattened above, rounded beneath, with dorsal surface outwards; then the base lengthens out, forming the tapering base or petiole of the spathulate leaf. In other cases the young leaves have the fully developed form from a very early stage.

Stem very soft and succulent, as is the whole seedling plant, with short internodes 1 mm. to 1.5 mm., but longer in the very early state, and continuing to lengthen after full development of leaf, branching early from the lower nodes; branches semi-erect, leafy; leaves as on main stem.

Variations: Often the lower part of the petiole is hardly adpressed to the stem, and the whole leaf is almost patent. The laminæ vary slightly in breadth. One seedling is stained conspicuously with pink in nearly all its parts.

Between the juvenile and mature plants, so far as. observed, there is no resemblance. The adult has much-reduced leaves, closely imbricating and adpressed to the stem, and in shape linear-triangular, with, very broad sheathing connate base, fleshy, plano-convex or slightly concave on the dorsal surface, ciliated, especially towards base, with rounded obtuse apex, and dotted with minute hairs on the ventral surface. Brought into cultivation, it quickly responds to the stimulus of moisture and shade, puts forth new growth, especially at the bases of the branches, which soon become furnished with numerous shoots very like young seedling plants, but with leaves not nearly so broad. Towards the apices of the branches the young growth shows many transitional forms, from the mature form, only slightly less imbricating, to linear leaves, patent and spreading, except towards

– 379 –

their bases (Plate XXXIV., figs. 56, 58, 57). The four apical leaves of each shoot in the seedling, as in most seedling New Zealand Veronicas, through their margins approaching and often touching, form a cup, which holds water and retains it for a considerable time. The young leaves also are readily wetted, the moisture being retained by the hairs, which may perhaps be organs of absorption. In general appearance this seedling form much resembles that of Saxifraga oppositifolia, L., also a Xerophyte, and growing in the high mountains of Europe.

No. 520. Veronica raoulii, Hook. f.

Seed gathered from plant growing on rocky face at gorge of Broken River by A. H. Cockayne. Seed germinated rapidly.

Description of Seedling.

Hypocotyle wiry, usually procumbent, variable in length, 1 cm. more or less.

Cotyledons rather fleshy, glabrous or furnished on petiole with very minute transparent. hairs, 6 mm. long; lamina entire, ovate, obtuse, green above, purple beneath or purple on both sides; petioles subterete, flat or slightly channelled above, connate at base.

Stem ascending, often bent towards extremity, terete, sparingly hairy with usually very short hairs; internodes variable in length (owing probably to the plants examined being much crowded).

1st pair of leaves ovate, crenate, at firsts 8 mm. long, faintly hairy with very minute hairs, especially on the petiole, dull-green on upper surface, purple on under-surface, with evident midrib; petioles as long, as or a little longer than lamina, subterete, faintly channelled above, connate at base.

2nd pair of leaves very similar to.1st pair, channel of petiole continuing half-way up lamina; margin of leaf slightly-recurved, especially towards obtuse apex.

V. raoulii grows on dry rocky faces, often exposed to full sunshine, at an elevation of from 600 m. to 900 m. I have never found any wild seedlings, and, although widely spread, it never occurs very abundantly. Formerly it was found on stony river-beds on the Canterbury Plains, according to Mr. T. W. Adams, and there also it would be exposed to very great drought. In cultivation in my shade - house plants from Mount Isabel, Hanmer Plains district, have reverted in their young growth to a semi-juvenile form with thin spathulate leaves, lobed at apex, and with two teeth on each side, with petiole equalling lamina—a very different form from the adult, with its thick, coriaceous, yellowish-green, linear-spathulate, short-petioled, serrate leaf.

– 380 –

No. 354. Veronica obovata, Kirk. Plate XXXII., fig. 23.

Seed collected from plant growing on bank of Craigieburn Creek, at 800 in. altitude. Germinated quickly.

Description of Seedling.

(Plant from which the measurements were taken was 3.5 cm. high.)

Root stout and woody at base, very deeply descending, brownish; lateral rootlets few, short.

Hypocotyle 6 mm. long, terete, pale-green, glabrous.

Cotyledons obovate; lamina 5 mm. × 3 mm., pale-green, entire, obscurely retuse, marked with a few most minute hairs, tapering slightly at base into petiole; petiole 2.5 mm. long, flat, connate at base. Leaves spreading, often patent, petiolate, subglaucous, much more crowded towards ends of branches.

1st pair of leaves 6.25 mm. × 4 mm., narrow-oboyate, subacute, entire, furnished with numerous most minute hairs, tapering into the petiole; petiole 5 mm. long; midrib swollen below towards base of leaves; margin slightly recurved.

2nd pair of leaves similar.

3rd to 6th pairs very similar to above, but sometimes marked with one or two rather deep serrations.

Later - developed leaves sometimes stained with red at apex and on teeth (when present), and with shorter petioles and broader laminæ.

10ch pair of leaves: Lamina from one seedling 1 cm. × 6 mm. and from another 1.1 cm. × 6 mm.; petioles 2 mm.

Stem soon branching from lower internodes, terete, pale, sometimes pinkish or bright-pink, extremely pubescent with short hooked hairs; internodes 3.5 mm. long; nodes slightly swollen, pink or purplish.

The seedlings examined did not vary to any great extent. Their growth appears to be slow.

Nos. 310, 634. Veronica epacridea., Hook. f. Plate XXXIV., figs. 59, 60, 65.

Seed from two sources—No. 310 gathered from plants on the western side of one of the rocky peaks of the Craigieburn Mountains, at altitude of 1,860 m.; and No. 634 from plant growing on shingle-slip on Mount Torlesse Range, at altitude of 1,000 m. Seed germinated in about four months.

Description of Seedling.

Root slender, deeply descending, with numerous side rootlets.

Hypocotyle one-half subterranean, 2.5 mm. long, terete, purplish on aerial part.

– 381 –

Cotyledons 3 mm. × 2 mm., fleshy, triangular - oblong, obtuse, glabrous, entire, with broad petioles; petioles connate at base, equalling lamina.

1st pair of leaves broadly triangular-ovate, obtuse, patent, curving downwards towards apex, sparingly hairy with minute hairs on upper surface, truncate or subtruncate at base; under - surface broadly keeled; petioles deeply and broadly channelled, connate at base, equalling lamina, which is 4 mm. × 3 mm.

2nd pair of leaves similar to 1st pair, but rather larger.

Later leaves (up to 8th pair) still of same type, but sides bending inwards, making upper surf ace concave; petioles from semi-patent to almost adpressed to stem.

Leaves on branches: These are considerably smaller and much more arched than leaves on main. stem, the petiole semi-erect and lamina. bent, arching downward almost at right angles, with apex recurved.

Stem very early in development branches from base, at first quite succulent and fleshy, more or less purple, sparingly minutely pubescent, terete; internodes 2 mm. long, but much shorter towards apex of stem.

Variations: The two batches of seedlings vary considerably: from one another: first, in colour—634 pale-green, only, later leaves faintly stained on margin with purple; 310-dark-green, edges of later leaves- very deeply stained with purple; second, in size of internodes, 310 having much larger interr-nodes; and, third, in size of leaves, the younger (634) having larger leaves than the older (310). The individuals of each, batch do not seem to vary to any extent, but not sufficient seeds have germinated to warrant any conclusion.

The whole plant is very succulent and soft. Such structure is an admirable provision against drought, growing as it does on solid rock or shingle-slips, for it cannot put down, a long root in search of water as the adult plant can; nor is. there so much danger of its drying, up with excessive trans, piration, since, being of very low stature, the large stones of the shingle-slips or the fissures in the rock, where alone the seed can germinate, will protect it from the drying winds. The. same remark would apply to V. tetrasticha, a companion plant. It is curious that this soft succulent form of leaf is the permanent form of V. haastinι, a closely allied plant, restricted to regions subject to the western rainfall.* V. epacridea has adult leaves (Plate XXXIV., fig. 65), extremely

[Footnote] *In the Handbook it is said to grow on Mount Torlesse; it must, however, be very rare; whereas in the position indicated above V. haastii grows with the greatest luxuriance, trailing in long patches over the stony ground. I think it most likely that the Torlesse plant was a. very open-leaved or a young form of V. epacridea.

– 382 –

coriaceous, more or less imbricating, subsessile, and in my cultivated plants ciliate towards the base. Hooker says glabrous.

Veronica pinguifolia, Hook. f. Plate XXXIV., figs. 61, 63, 64.

(Var. with prostrate stems, rooting freely from the nodes.) Seed from cultivated plant in Tarata Garden.

Description of Seedling.

Root remains enclosed in seed for considerable period; in one case the stem was furnished with two pairs of leaves before it emerged, at first Quite coiled up, finally descending, with very many filiform lateral rootlets.

Hypocotyle 5 mm. long, terete, green or pinkish.

Cotyledons 5 mm. long, ovate-oblong, obtuse, green above, sometimes purplish on under-surface; midrib hardly evident; petioles connate at base.

Leaves pale-green, pinkish, or margin more or less deeply stained with pink; topmost crowded; earlier leaves with petioles semi-erect, and laminæ spreading and curving downwards often, usually narrow- or broadly-ovate, obtuse, with petioles from one-third to one-half length of, leaves; more mature leaves broader, almost rotund at times; midrib slightly raised on under-surface; both surfaces dotted with many minute white scales; still later leaves obovate, almost sessile, quite glaucous above, with a red margin.

1st pair of leaves narrow-oblong or obovate, obtuse, green on upper sometimes purplish on under surface; midrib partly evident; petioles connate at base.

Adult leaf (Plate XXXIV., fig. 63) coriaceous, 1.3 cm. × 7.5 mm., sessile, glaucous, patent, sometimes twisted towards light; margin entire, pale, slightly stained with pink; surface concave above, convex below; midrib sunken below.

Stem green at first, then pink or brownish; internodes short, about 2 mm. long, do not. seem to lengthen, almost glabrous, with a very few scattered hairs and scales.

The variations among the seedlings seem to be slight; one has a 2nd leaf with an emarginate apex, and there is some variation in breadth of leaves and in the pink stain. The variations will be more marked at a later stage. V. pinguifolia is essentially a plant of the drier mountains, xerophilous in structure, and not occurring, so far as I know, on the western side of the dividing-range, and possibly not within the region of the western rainfall.

No. 661. Veronica linifolia, Hook. f. Plate XXXIV., figs. 52, 53.

Seed gathered from plants growing in shade on wet rocks,

– 383 –

Waterfall Creek, Craigieburn Mountains, by A. H. Cockayne, at altitude of 700 m. Seed germinated in about six weeks.

Description of Seedling.

Plants very small, with stem hidden by the close-set patent shining leaves.

Root descending, fleshy, with very few lateral rootlets.

Cotyledons ovate, 2mm. long, acute, brownish-green above; pink beneath, furnished with many minute scales; petioles short, subterete, flat or slightly channelled above.

Leaves broadly spathulate, extremely succulent, green, subacute, scaly as cotyledon, patent; petiole two-thirds length of lamina, translucent, subterete, channelled above, connate at base, ciliated on margin with a few transparent hairs.

Stem very succulent, translucent, terete; internodes very short (in plant with three pairs of leaves), 0.75 mm. or less.

Veronica linifolia grows on wet rocks, often in almost perfect shade, in moist river-gorges. The succulent seedling seems rather out of place under such circumstances until one bears in mind that even in a region so wet as the western side of the Southern Alps a drought of a few days makes the ground wonderfully dry, and a plant exposed normally to much wet would suffer extremely unless provided with a water-supply. Figs. 52 and 53, Plate XXXIV., show the differences between the juvenile and adult leaves.

Veronica macroura, Hook. f. Plate XXXIII., figs. 36, 37, 38, 39, 40, 41, 42.

The plant from whose seed the seedlings described below were raised was kindly sent to me some years ago by Mr. T. Kirk, F.L.S., under that name; its leaves, however, do not agree with the description in the “Handbook of the New Zealand Flora,” page 207, being ovate or ovate-oblong, almost or quite sessile, and rounded at the base, subapiculate, minutely ciliated; but the slender pubescent slightly curved racemes agree with Hooker's description. Whether it be the true V. macroura or not, it reproduces itself quite true from seed, and is most certainly a distinct species.

Description of Seedling.

(Drawn up from self-sown seedling plants in Tarata Garden.)

Root wiry, giving off numerous, rootlets equalling the primary root.

Hypocotyle short, terete, minutely pubescent.

Cotyledons. oblong, fleshy, entire, obtuse, dotted with minute scales; laminæ 3 mm. × 2 mm.; petioles 2 mm: long, flat, connate at base.

– 384 –

1st pair of leaves fleshy, tender, rotund, rotund-ovate or ovate, obtuse; midrib sunken above, hardly evident beneath; lamina 6 mm. × 5 mm., minutely ciliated, and with a few white scales on the surface; petioles 3 mm. long, subterete, slightly channelled, connate at base.

2nd pair of leaves obovate, entire or lpbed, more distinctly ciliated; lamina tapering into petiole, otherwise as first leaves.

3rd pair of leaves similar, but larger, and margin crenate above middle of leaf.

Further development of leaves: The later leaves (Plate XXXIII., figs. 36, 40, 41) become almost sessile, with white midrib above, less distinct beneath, with margin crenate above centre, with shallow lobe at apex, and with base subcuneate. These leaves are followed by larger leaves less cuneate at base; these by almost oval sessile leaves, still crenate on upper half of sometimes red-stained margin; and later on still subacute obovate leaves appear.

Stem at first erect, then apex inclines downwards and horizontal branches are developed from the early nodes, terete, at first pinkish, afterwards pale-green, hoary with grey pubescence; internodes (in plant 8 cm. high) 4.5 mm., long.

The development as described above does not as yet show the change into the very distinct adult form (Plate XXXIII., fig. 42).

Veronica diosmæfolia, R. Cunn. Plate XXXII., fig. 26.

Seed gathered from cultivated plant in garden of Mr. W. Martin, Fairfield, Dunedin. Germinated rapidly, and plants are of extremely fast growth.

Description-of Seedling.

Root long, wiry, and with numerous long lateral rootlets, especially from base.

Hypocptyle very short, 2 mm. or 3 mm. long, finally very stout and. woody.

Cotyledons oblong, entire, slightly emarginate, succulent, pale-green, very short petioled; petioles connate at base.

1st to 3rd pairs of leaves lanceolate, upper half of margin deeply toothed, acute, sometimes dotted on both surfaces with minute white scales, tapering into the petiole; petiole subrterete, channelled above, one-quarter length of lamina; lamina 6 mm. × 3–5 mm.

4th or 5th pairs of leaves and upwards most uniform in shape, linear or very narrow linear-lanceolate, almost sessile, convex on upper surface owing to infolding of sides of lamina; midrib evident on upper and under surfaces; margins serrate

– 385 –

above middle, with usually four rather distant teeth; laminæ inclined a little from the horizontal to the vertical.

Stem brown, branching at an angle of about 45°, alternately bifariously pubescent; internodes 3 mm. long; nodes swollen and purple, especially at junction, with petiole.

No. 345. Coprosma acerosa, A. Cunn. Plate XXXIV., figs. 69, 72.

(Erect var., as found in mountainous situations.)

Seed collected from plant growing amongst other shrubs at base of Mount Earnslaw, Lake Wakatipu. Germination very slow; some of the seeds did not germinate until twelve months from date of sowing.

Description of Seedling.

Root straight, deeply descending, with a few yellow lateral rootlets.

Hypocotyle 4 mm. long, becoming rather longer as plant developes.

Cotyledons 5.5 mm. × 2 mm., narrow linear-oblong, slightly widest near base, obtuse, patent, rather thick and coriaceous; petioles very short; midrib evident below. As the plant developes the cotyledons increase. in size, and are persistent until the 6th or 7th leaves appear.

Stem woody, terete, swollen at, nodes, with internodes, in plant 4 cm. high, 1st 5 mm. long and 4th 9 mm. long, pale yellowish-brown dotted with red or purple, pubescent everywhere, with numerous very short hairs.

1st pair of leaves 7 mm. or 8mm. × 2.5mm, linear-oblong, obtuse, green, brownish on midrib and margin or entirely brownish on upper surface, which has numerous minute scales and a few white hairs in fascicles, paler and scaly on under-surface, obtuse; margins minutely serrulate; midrib evident.

Succeeding leaves much same as 1st pair, often concave above, patent and bending downwards for upper half, but the later leaves are acute, with often a triangular apex, the margins below being quite parallel, green or stained with brown as before.

Stipules: Two, surrounding leaf-bases, interfoliar, connate, triangular, hairy, with from one to three glandular swellings at apex (Plate XXXIV., figs. 72, 69).

Further development: From each axil in the most developed plants a pair of leaves are being developed at right angles to the axis of the earlier leaves.

Variations: The seedlings vary slightly in the amount of pubescence, the purple-stemmed being the most pubescent. In some plants there are slighter broader leaves than those described, and one is subcordate at the base.

– 386 –

The extremely fleshy stipules found in this and all other Coprosmas which I have examined, with their glandular swollen tips, seem to be organs of protection for the very young bud, any injury to which would very much check the growth of the plant. Seedlings in the wild state are found most abundantly under the shelter of bushes, especially where there is running water and the adult plants are either constituents of subalpine scrub or occur with other shrubs usually on the shady side of river-terraces. Coprosma acerosa, of sand-dunes, and the prostrate river-bed variety, with blue or white fruit, heed studying in the seedling form, when it will be found, I think, that they are distinct species. The latter I raised from seed some years ago, and it reproduced itself, but my notes are not full enough for publication.

Coprosma petriei, Cheeseman.

(Var. with white fruit.)

Seed gathered from plants growing on stony fiat at Glenorchy, Lake Wakatipu. Germination very slow, but seedlings numerous.

Description of Seedling.

Root very long, 7.5 cm. in plant examined, straight, deeply descending, yellow; rootlets very few, extremely short.

Hypocotyle 4—5 mm. long, terete, suffruticose, lengthening much as development proceeds.

Cotyledons linear, entire, obtuse, covered on upper surface with, white adpressed scales, and on under-surface with minute protuberances, gradually narrowing into the short, broad, connate petioles, 8 mm. × 2.5 mm.; soon withers.

Leaves opposite, cauline, stipulate, short-petioled, variable in shape at all stages but all of one type, most common linear-oblong, sometimes narrow-oval, spathulate or linear-obovate, entire, green, thick, fleshy, acute or obtuse, covered on both surfaces with a thin translucent scaly brittle deciduous pellicle, beneath which are abundant-minute scaly hairs; midrib obscure, white on under-surface of leaves.

1st pair of leaves 6 mm. × 2 mm., linear-oblong; petiole rather more than half length of leaf.

2hd pair of leaves narrow obovate-spathulate, 9 mm. × 3 mm.; petiole half length of lamina.

Stem pale dotted with pink, terete, swollen at internodes, where it is surrounded by two interfoliar stipules; 1st inter-node 3 mm., 2nd.internbde 2 mm. The stem branches laterally from axils of leaves, bearing leaves similar to those described above, except these are. very distinctly ciliated on margin with white hairs, and when first developed are spathulate or linear-obovate.

– 387 –

Stipules interfoliar, adhate with, petioles, triangular or truncate, furnished with three very short deciduous red glandular hairs at apex, ciliated, rather fleshy at first.

The adult leaves differ considerably from those described above, being narrow-lanceolate, extremely acute, long-petioled, and furnished at regular intervals on both surfaces with stout, acute, erect, white hairs; midrib, margin of leaf, and petioles deeply stained with reddish-brown purple. The seedlings appear to vary very little. C. petriei grows in full sunshine, closely pressed to the ground, forming a turf, on old stony river-flats or on stony river-terraces where there is very little earth and exposed to great heat in summer time, at altitudes of from, about 300 m. to 700 m., and covering large tracts of ground.

Coprosma areolata, Cheeseman.

Seed gathered from plant at base of Signal Hill, Dunedin. Sown seven months after gathering. Germinated in abut ten months.

Description of Seedling.

Root: Primary root straight, descending, very stout, with numerous extremely slender filiform lateral rootlets.

Hypocotyle 1cm. long, terete, tapering dotted with purple, glabrous.

Cotyledons oblong, obtuse, dotted with minute scales on under-surface, pale-green, sometimes later on becoming purple-stained, 7 mm. × 3.5 mm. to 9 mm. × 4 mm.; petioles short; midrib evident, pinkish; margins entire, stained with light-purple.

1st pair of leaves: At first almost in same plane with cotyledons they become gradually raised by lengthening of petioles, while at the same time the apex becomes more acute, and points upwards; 13 mm. × 6 mm., ovate, acute, entire, With red margins and veins, ciliated on margin and midrib; upper surface pale apple-green; under-surface. paler than upper; veins much anastomosing; midrib raised on under-surface, especially towards petiole; petiole semi-terete, channelled.

Later leaves very similar; to above, but larger; base of leaf tapering into and decurrent with upper part of petiole; upper surfaces, petioles; and margins hirsute with many long strict, acute, white hairs; upper, surfaces of leaves often marked with purple blotches, as indicated in younger leaves by pinkish-purple margins, &c., as described above.

Stem at first very short; cotyledons and 1st pair of leaves appearing to emerge from same spot, gradually lengthening as the 1st pair of leaves develope; then erect, freely budding

– 388 –

and branching from axils of leaves; internodes lengthening considerably as growth proceeds—in plant 5.3 cm. high, 1.1 cm, to 11 mm. long, hairy with short dense pubescence, and also with numerous long acicular white hairs; hairs 1.5 mm. to 2 mm. long, strict, pointing at right angles to axis of stem.

Stipules 2, interpetiolary, adnate, with petioles from broadly triangular to very broadly rectangular, - hairy, and furnished at apex with one long fleshy purple mucro.

No. 387. Aristotelia fruticosa, Hook. f. Plate XXXI., figs. 16–20a1h, and Plate XXXIV., fig. 73.

(Var. with white fruit.)

Seed gathered from plant on Ben Lomond, Lake Wakatipu, at altitude of 400 m. Germination was very slow.

Description of Seedling.

Root brown, flexuous or straight, furnished abundantly with short filiform lateral rootlets, often also with large spreading stout rootlets from base of hypocotyl, nearly equalling the primary root.

Hypocotyle succulent at first, soon stout and woody, finally about 1.5 cm. long, glabrous below, hairy above with adpressed brown hairs.

Cotyledons leafy, persistent for long time, still green on plant 9 cm. high, 8 mm. × 4.5 mm., varying in shape and size, oblong, obovate or oval, obtuse, in one instance retuse, furnished with many minute scales, especially on under-surface, hairy with brown adpressed hairs on the petiole; margins entire, faintly ciliated, especially towards petiole; upper surface dull dark-brownish or yellowish-green; under-surface paler and redder; petioles short, plano-convex, connate at base.

Leaves extremely variable at all stages, even up to the full development of the shrub; early forms of very thin texture, drying up rapidly if exposed to drought, green, brown, blackish-green or reddish-brown on upper surface; of similar tints but paler and more shining on under-surface; shape most variable; first few leaves usually ovate to lanceolate (Plate XXXI., figs. 14, 18); later leaves linear-lanceolate to broad-lanceolate, at times narrow-triangular (Plate XXXI., figs. 15, 19, 20, and Plate XXXIV., fig 73); still later leaves* (Plate XXXI., figs. 20a, b, g, h) obovate, obovate-oblong, cuneate at the base, entire, slightly or much serrate, with intermediate most curious forms by reversion, linear-lanceolate, serrate or pinnatifid, with segments linear or linear-oblong,

[Footnote] * From wild seedling in cultivation for eight months.

– 389 –

sometimes pectinate; serrations and sides of leave often much incurved; colour of later leaves often almost black, very dark-brown or dull-green in centre, surrounded by dark margin; margins (of earlier leaves) rarely regularly serrate, usually irregularly biserrate, ciliated; midrib much raised on under-surface, hairy; venation penninerved and reticulating, with veins often much swollen on under-surface of leaf; petioles subterete, variable in length, slightly channelled; connate at base.

Size of leaves: This may be seen from the figures in the plates, which are natural size. In the older plants some of the leaves are considerably larger than those figured.

Stem (in young seedlings), erect, straight, usually pink in early seedling, in older seedlings much darker, almost black at times, with very dark-purple blotches, pubescent with short straight hairs sometimes curving upwards at the tip; inter-nodes averaging about 8 mm. long in plants 6 cm. high; in older seedlings (the tallest 24 cm. high) are stout, straight, horizontal branches at right angles to main axis, sometimes opposite, sometimes alternate, with leaves like those on main axis opposite and sometimes with surface quite flat and exactly horizontal.

Final development into, mature state with more simple reduced leaves (Plate XXXI., figs. 16, 17) not yet observed.

The various forms assumed by the leaves of A. fruticosa are almost beyond belief; still a regular sequence of forms from the early thin-leaved ovate to the later coriaceous obovate by way of all varieties of lanceolate can be traced. Perhaps the most remarkable of all are the narrow lanceolate or triangular forms with truncate bases (Plate XXXIV., fig. 73), observed in the collected seedlings, which were reverting to the lanceolate early seedling form (Plate XXXI., fig. 15). At this stage the leaves often assume a considerable size (Plate XXX.I:, fig. 20), and in many instances are incised almost to the midrib. The early red-stemmed ovate-lanceolate leaved seedling much resembles A. racemosa and A. colensoi in miniature, and these latter may be looked upon as plants arising from a common ancestral stock which have kept almost intact the ancestral character, whereas in A. fruticosa this has been entirely changed by its subalpine environment. It is true that A. colensoi and A. racemosa also at times reach the subalpine zone, but it is in places where they have abundant shelter and moisture. The final form of A. fruticosa becomes very similar to that of its companion plants referred to before when treating of Pittosporum rigidum; and, with its dense rigid tortuous branches and small coriaceous leaves, it might well be taken for A. coprosma. In the wild state very

– 390 –

many adult forms, of A. fruticosa are met with, many of which must depend upon the local environment, and which will be possibly of the most unstable character.

No. 389. Plagianthus divaricatus, Forst. Plate XXXI., figs. 91 10, 11.

Seed collected from plant growing in salt neadow near New Brighton, Canterbury. Germination very rapid, from eight to fourteen days.

Description of Seedling.

Root fleshy, straight, tapering, descending, pale, with, few very short lateral rootlets.

Hypocotyle terete, glabrous, tapering upwards from the thickened base, slightly twisted or bent, white below, pale-green above.

Cotyledons from subcordate to almost rotund, 3- or 5-nerved, with much reticulated venation, sometimes slightly lobed or angled towards apex, minutely glandular-pubescent; petioles subterete, slightly channelled, almost equalling the lamina.

Leaves alternate, stipulate.

1st leaf broadly ovate, almost glabrous, entire or marked with slight notch towards the somewhat abrupt narrowing of the apex; petioles channelled, half length of lamina.

2nd leaf oblong to broadly oblong; margins marked with one or two crenations, forming three irregular lobes.

3rd leaf similar in shape, but more narrow than 2nd leaf.

Later development of leaves: The succeeding leaves become more narrow, for a time each leaf a little narrower than its predecessor, until (Plate XXXI., fig. 11) the usual form is linear-lanceolate with petiole half length of lamina, pale-green, nerves faintly swollen on upper surface, and sometimes falcate in shape.

Stem at first wiry, brown, terete and straight, then becoming flexuose and extremely pliant as development proceeds, still very slender, especially, towards slightly drooping extremity; hairy with minute white stellate hairs; thickened at nodes, whence arises one oblong or triangular, truncate at apex, very hairy stipule; branching at rather distant intervals with long divaricating branches similar to. main axis, also with Very short branches -with suppressed internodes and leaves in fascicles.

P. divaricatus is finally a rounded, most dense shrub, with interlacing, short, stout, pliant, much-divaricating branches bearing many small linear or obovate-lihear falcate sessile leaves, usually in fascicles. On the outside of the shrub these leaves are extremely fleshy, and often have the dorsal surface

– 391 –

turned to the incident light, but inside and sheltered are-much thinner. Younger plants have the upper twigs.erect, not interwoven, and have stellate hairs, as in the seedling form, but extremely reduced fleshy linear leaves. It grows in Wet salt meadows, near the sea-shore, and is exposed to most frequent east winds, and when these do not blow to heavy gales from north-west and south-west. This constant exposure to wind accounts for the final habit of growth; which much resembles that of Coprosma acerosdy, growing on the adjoining sand-dunes. It is interesting; in connection with the above chat P. linariifolia, Buchanan, of the West Coast, from moister and probably more sheltered habitat, is, judging from his drawing and description (Trans. N.Z. Inst;, vol. xvi., pl. xxxiv., pp. 394, 395), much more like the later seedling form of P. divaricatus than is its own adult form.

Pseudopanax crassifolia, C. Koch. Plate XXXIII., figs. 31–35.

Described from seedlings collected on bank of River Kowai, Mount Torlesse Range, and which have been Cultivated in greenhouse under same conditions as other seedlings treated of for eight months.

Description of Seedling.

Root long, stout, often with many lateral spreading rootlets from the base; young rootlet with great numbers of root-hairs.

Hypocotyle variable in length, thick, fleshy at first, then woody, dark-brown or dark-purple, glabrous, smooth, terete.

Cotyledons, persistent, leafy, obovate (fig. 33), oblong, entire or toothed, rounded obtuse or very shallow-lobed at apex; midrib prominent, swollen; margin pale-reddish slightly recurved; petioles short, semi-erect, connate at base.

1st leaf coriaceous, dark-green, often blotched with pale-brown, ovate-lanceolate, cuneate at the base or rhomboid, deeply and coarsely toothed, tapering gradually into the petiole; petiole semi-terete, channelled above, half length of lamina, sheathing and swollen at the expanded base.

2nd leaf usually linear-lanceolate, sharply toothed; teeth largest towards base of leaf; apex acute surfaces—upper, black-green, spotted with pale-brown;. under, much paler, tapering into the petiole: midrib evident on both surfaces, and raised; petiole semi-terete, slender, channelled, much shorter than lamina, sheathing and swollen at base.

3rd leaf often very similar to 2nd, or already the second type of leaf may have appeared as under.

4th and next few succeeding leaves sessile, or almost so, spreading and pointing slightly upwards, linear, with distant

– 392 –

regular alternate short stout teeth (quite different from the coarse serrations of the first type of leaf), with yellowish rather blunt tips and their bases on upper surface of leaf slightly swollen; upper surface very dark-green, often marked with pale-brown blotches, especially along midrib and bases of teeth; under surface paler than upper, and brown on midrib and round margin; midrib very prominent, raised and reddish-yellow on upper surface; petiole or base of leaf expanding into amplexicaul sheath slightly swollen.

Stipules subulate, membranous, adnate to leaf-sheath at base.

Stem erect, often bent or twisted, stout, woody, brownish, marked with many oblong or round white blotches; inter-nodes very much reduced above, longer below; several of the seedlings under examination 4.5 cm. high are already branching from near apex of primary stem.

The further development has been treated of at considerable length by Kirk (“Forest Flora,” pp. 59, 60; and Trans. N.Z. Inst., vol. x., Appendix, pp. xxxi. to xxxiv.). He shows that there are various varieties of P. crassifolium, and that some of these have distinct and constant seedling forms, while one, the form of the Chatham Islands, never passes through the long-leaved seedling form of which my description above is but a preliminary stage. If there are several seedling forms distinct from one another except in the very earliest stage, then I should certainly think, from biological grounds, that there are as many species as there are distinct seedling forms. The matter is certainly one which needs going into again from a specific point of view, to say nothing of the immense interest attached to any facts that might be elicited in regard to the conditions which cause the various leaf-changes. It would, I think, be easy to cause an adult to revert to the juvenile form simply by cutting down a tree or removing a considerable limb, when the new growth would be the juvenile most likely. Cuttings of the mature form could also be struck, and, when rooted, experimented with under various known conditions. Botanists in the North Island, where the var. trifoliata is to be found, might easily undertake this interesting work. Kirk says, “The stem is invariably simple in this stage of development, ranging from 6 ft. co 20 ft. in height before branching”; but in my seedlings, grown in moist greenhouse, &c., they are already branching, as stated above.

No. 326. Olearia odorata, Petrie. Plate XXXIII., fig. 43.

Seed collected from shrub growing in Waitaki Valley by Mr. Rutherfurd, Rugged Ridges. Germination slow.

– 393 –

Description of Seedling.

Root extremely stout, tapering, deeply descending, with lateral rootlets mostly from upper half.

Hypocotyle very short, subterranean, hardly to be distinguished from the root.

Cotyledons wither very early, rotund, petiolate. (Too much withered for further description.)

Leaves at first crowded together, appearing almost radical through the slow development of the internodes, afterwards becoming much more distant; appear to increase considerably with growth of plant for some time; opposite, short-petioled, spreading, patent, from ovate-oblong to spathulate, pale whitish-green, entire, in one specimen slightly lobed, furnished with many minute scales on both surfaces, and with numerous adpressed hairs on petioles, midribs, and. bases of laminæ, but more sparingly hairy on surfaces and margins; laminæ narrowing very gradually into petioles; petioles partly sheathing, and connate at bases. The more early developed leaves are much broader than the later ones. Measurements—-Early leaves, 2.5 cm. × 1cm.; later leaves, 1.10 cm. × 5.5 mm.

Stem erect, terete, stiff, rather densely covered with adpressed white hairs, purple and swollen at the nodes, forming leaf-scars when leaves are shed; internodes, in plant 5.4 cm. high, 11 mm, long.

Further development not yet observed.

No. 427. Gunnera dentata, Kirk. Plate XXXIV., fig. 48.

Seed gathered from plants growing on very wet clay bank, near Glenorchy, Lake Wakatipu. Germination very slow. Very few seeds germinated, and seedlings of extremely slow growth.

Description of Seedling.

Root thick, fleshy, rather short; afterwards from stem appear fusiform adventitious rootlets with bright, shining, crimson tip, and furnished with very many extremely slender root-hairs near the base, these at first aerial.

Hypocotyle short, semi-prostrate, very pale-green, sparsely hairy with straggly hairs.

Cotyledons very pale-green, 6 0mm. × 2 mm. to 4 mm. × 2 mm., spathulate, extremely fleshy and succulent, with a few long pale hairs and many minute scales, convex on upper flat on under surface, spreading, with surface almost horizontal.

1st leaf 5 mm. × 3.5 mm., almost rotund in outline, fleshy, truncate at base, entire for lower half of margin, and coarsely serrate, with two opposite teeth on upper half, and sparingly hairy; apex lobed, acute; upper surface of

– 394 –

lamina marked with numerous minute scales and with a few straggling long twisted hairs; under-surface of lamina with stout midrib and indistinct veins, more hairy than upper surface; petioles very, stout, sheathing and connate at base, brownish-green, pilose.

2nd leaf rounded at base, more toothed than 1st leaf.

3rd, 4th, and 5th leayes often more narrow than the 1st and 2nd, obovate or oblong, with toothing less coarse and apices of teeth thickened, lamina sometimes tapering at base into the petiole, petiole as long or nearly twice as long as lamina, channelled above, sheathing beneath.

Stem (in plants with, five leaves, the largest examined) very short, thick and fleshy, sometimes with internodes almost suppressed, giving off adventitious roots, which first appear as crimson protruberances, hairy as leaves.

Further development not yet observed.

The material from which the above was drawn up was. not good, and so the description may not be strictly accurate.; also, details of interest, may have been left out. It is possible, the earliest leaves may sometimes be much narrower and less toothed than the one figured. The species grows most luxuriantly on banks of sluggish streams in mountainous districts, also on wet banks or on rock in the drip of water. The young plants examined had extensive colonies of Nostoc in the parenchyma of the stem just at the lower-leaf bases,* Between the adult and juvenile forms there seem to be no very great differences.

No. 416. Hymenanthera dentata, R. Br., var. angustifolia, Benth. Plate XXXIII., figs. 44–47.

Seed. collected at Kingston, Lake Wakatipu. I am not sure of the correctness of my identification, not having seen the type specimens in Kirk's herbarium. It is, however, one of the forms called by Hooker, in the “Handbook of the New Zealand Flofca,” H. crassifolia, and it is also the plant under that name in Petrie's “List of Otago Plants” (Trans. N.Z. Inst., vol. xxviii., p. 543).

Description of Seedling.

Root long, descending deeply, thick, with many lateral rootlets.

Hypocotyle: One half subterranean, very fleshy at first, then woody, pale-coloured, finally pinkish, often bent, terete, glabrous.

[Footnote] *I am indebted to Dr. K. Goebel for showing me Nostoc in G. monoica. and so calling attention to the fact that New Zealand Gunnera, as welt as South American, possess Nostoc colonies.

– 395 –

Cotyledons 1.1 cm. × 7.5 nun., oblong, oval or rotund-oblong, entire, retuse or emarginate, rarely. obtuse, thick coriaceous, very persistent, dull, dark-green, with many, minute scales, especially on the under-surface, 3-nerved from base; midrib on entrance into leaf almost immediately giving off two more nerves, which are parallel for a time, and then converge, towards apex of leaf; petiole very short, subterete, flattened above, 2 mm. long.

Early leaves alternate, stipulate, variable in shape, often from oblong to linear-oblong, blackish-green on upper paler on under surface, short-petioled or sessile; midrib and-veins swollen on upper surface, indistinct on under-surface, dotted with many minute white scales; margins entire or serrate,; serrate margins with serration on one side, with one opposite serration on each side, usually nearer base than centre; apex lobed to subacute.

Later leaves pale-green at first, linear to linear-oblong, tapering into very short petiole, smaller than earlier leaves on main stem, usually entire, with very short apiculus at end of midrib, sometimes lobed and irregularly toothed as in early leaves; veins swollen on upper surface; laminæ at angle of 45° to axis of stem, semi-vertical to horizontal.

Stipules very small, one on each side of petiole, subulate.

Stem of slow growth, very stiff, erect, and woody, early branching with stiff divaricating branches, rather pale-brown, with many flat white scales at regular intervals, younger portions of branches with two rows of very short close-set hairs, terete; internodes 3 mm. long, more or less.

Further development not observed.

In one or two instances, the laminæ of the cotyledons were, partly or very nearly cleft to the base, showing how a plant may be developed with more than the normal two cotyledons. The toothing on the leaves. seems due to non-development of the nerve and accompanying portion of lamina rather than to ordinary, toothing, the two lateral nerves of a trinerved leaf ending in this case in the serrations, the midrib alone reaching the upper portion of the leaf. Hymenanthera dentata is another of the shrubby plants which grow under conditions of drought and wind, and its final adult stage resembles that of Coprosma acerosa, &c., treated of before.

No; 665. Celmisia bellidioides, Hook. f.

Seed collected by A. H. Cockayne at Waterfall Greek, 1 Craigieburn Mountains, at altitude of 1,200 m., growing on wet rock, in shade. Germination slow.

Description of Seedling.

Hypocotyle 6 mm. long, brittle, rather thick.

– 396 –

Cotyledons 5 mm. × 2.75 mm., linear-oblong, obtuse, sessile, unequal in size, thick, coriaceous, glabrous.

Leaves imbricating, with broad sheathing bases, matted together at base with long cobwebby white hairs.

3rd leaf: Lamina narrow-oblong, 5 mm. × 3.5 mm., one tooth on one side, rest of margin entire, obtuse, varnishes green on upper surface, paler on under-surface, tapering into the long petiole; petiole 5 mm. long, broader at base, sheathing and imbricating with other petioles.

Next few leaves very similar.

Further development not yet observed.

It is only the extreme difficulty of raising such a Celnisia as this from seed, and the further difficulty of keeping the seedlings alive when raised, that decided me to give the above meagre description.

Veronica armstrongii, Hook. f.

The seedling plant referred to in Part I., and of which a photograph is appended (Plates XXVIII. and XXIX.), was, for purpose of photographing, taken out of its small pot, and then repotted in a larger one, in the middle of October, 1898. The plant was kept until the middle of December in the greenhouse, along with the other seedlings, by which time the tips of the shoots in nearly every instance were beginning, to approach nearer to the adult scaly adpressed form. The plant was then placed under a bell-glass and kept very moist to see if under such conditions it would again revert to the seedling form. The experiment has been eminently successful, and now—February, 1899—every shoot is well furnished at the tip with seedling leaves, which are 4 mm. long, petiolate, linear-lanceolate, deeply toothed., with two opposite teeth on each side, one tooth or the older leaves entire, bright-green, soft and succulent; petioles broad, connate; midrib evident. The smaller more-reduced leaves further down the stem; have also opened out considerably, and their upper portion is quite patent. The plant is in good health, and growing vigorously. The shoots are leafy and green for their whole length, and the plant in its present state much more resembles V. loganioides, Armstg., than its own adult form. I am going to keep it under the same conditions, to see if it is not possible to bloom it while in this still juvenile form. The plant is now 13.6 cm. tall and 10 cm. in diameter, with branches opposite and quadrifarious.

Explanation of Plates XXVIII.–XXXIV.

Plate XXVIII.

Branch of Veronica armstrongii, T. Kirk. Photographed from cultivated specimen.

– 397 –

Plate XXIX.

Seedling plant of Veronica armstrongii, two years old, and portion of shoot cut from, same.

Plate XXX.

Fig. 1. Later leaf of seedling Carmichœlia gracilis.

Fig. 2. Seedling plant of C. gracilis.

Fig. 3. Seedling plant of Veronica salicifolia.

Fig. 3a. Third seedling leaf of V. salicifolia.

Fig. 3b. First seedling leaf of V. salicifolia.

Fig. 4. Seedling plant of Pittosporum rigidum.

Figs. 6. 5a. Later leaves of Carmichœlia robusta.(?).

Fig. 6. Leafy cladode of Carmichœlia hookeri.

Fig. 7. Seedling plant of C. hookeri, showing the later leaves.

Fig. 8. Early seedling form of C. hookeri.

Fig. 9. Early seedling form of C. robusta (?).

Plate XXXI.

Fig. 91. Seedling plant of Plagianthus divaricatus

Fig. 10. Cotyledon of P. divaricatus.

Fig. 11. Later seedling leaf of P. divaricatus.

Fig. 12. Seedling plant of Carmichœlia crassicaule.

Fig. 13. Portion of leafy cladode from adult C. crassicaule developed in cultivation.

Fig. 14. Seedling plant of Aristotelia fruticosa (cultivated).

Fig. 15. Later seedling leaf of A. fruticosa (cultivated).

Figs. 16, 17. Leaves from adult A. fruticosa.

Fig. 18. Early seedling leaf of A. fruticosa (cultivated).

Fig. 19. Later seedling leaf of A. fruticosa (cultivated).

Fig. 20 (a1 to h). Various forms of seedling leaves of A. fruticosa from wild seedling plants.

Plate XXXII.

Fig. 21. Seedling plant of Rubus australis, var. glaber.

Fig. 22. Seedling plant of Pseudopanax crassifolia (wild).

Fig. 23. Seedling plant of Veronica obovata.

Fig. 24. Seedling plant of Carmichœlia odorata.

Fig. 25. Seedling plant of Sophora microphylla.

Fig. 26. Seedling plant of Veronica diosmœfolia.

Fig. 27. Seedling plant of Veronica traversii.

Fig. 28. Early seedling form of V. traversii.

Plate XXXIII.

Fig. 30. Leafy cladode of seedling Carmichœlia robusta (?).

Fig. 31. Fifth seedling leaf of Pseudopanax crassifolia.

Fig. 32. Seedling plant of P. crassifolia, showing cotyledons.

Fig. 33. Cotyledon of P. crassifolia.

Fig. 34. Second leaf of P. crassifolia.

Fig. 35. First leaf of P. crassifolia.

Fig. 36. Later Seedling leaf of Veronica macroura.

Fig. 37. Seedling plant of V. macroura.

Fig. 38. Second or third leaf of seedling V. macroura.

Fig. 39. First or second seedling leaf of V. macroura.

Figs. 40, 41. Later leaves than No. 36 of V. macroura.

Fig. 42. Adult leaf of V. macroura.

Fig. 43. Seedling plant of Olearia odorata.

Fig. 44. Seedling plant of Hymenanthera, species.

Figs. 45, 46. Early seedling leaves of Hymenanthera.

Fig. 47. Later seedling leaf of Hymenanthera.

– 398 –
Plate XXXIV.

Fig. 48. First seedling leaf of Gunnera dentata, × 6.

Fig. 50. First seedling leaf of Ozothamnus depressus, × 6.

Fig. 51. Cotyledon of Ozothamnus depressus, × 6.

Fig. 52. Adult leaf of Veronica linifolia, × 1.

Fig. 53. First seedling leaf of V. linifolia, × 6.

Fig. 55. Early leaf of Veronica tetrasticha, × 6.

Fig. 56. Adult leaf of V. tetrasticha, slightly changed by moist shady atmosphere, × 4.

Fig. 57. Adult leaf of V. tetrasticha, almost reverted to seedling form, × 4.

Fig. 58. Adult leaf of V. tetrasticha, intermediate between Nos. 56 and 57.

Fig. 59. Later seedling leaf of V. epacridea, × 6.

Fig. 60. Second seedling leaf of V. epacridea., x.6.

Fig. 61. Second leaf of Veronica pinguifolia, × 6.

Fig. 62. First leaf of Veronica traversii, × 4.

Fig. 63. Adult leaf of V. pinguifolia, × 1.

Fig. 64. Late seedling leaf of V. pinguifolia, × 1.

Fig. 65. Adult leaf of V. epacridea, flattened out, × 1.

Fig. 66. Emarginate (abnormal) cotyledon of V. traversii, × 6.

Fig. 67. Glandular hair of Rubus australis, × 268.

Fig. 68. Hairs of Rubus australis, x. 268.

Fig. 69. Glandular, tip of stipule of Coprosma acerosa, × 50.

Fig. 70. Fifth seedling leaf of Sophora (Chatham Island var.), × 1.

Fig. 71. Fifth seedling leaf of Sophora grandiflora, × 1.

Fig. 72. Stipule, of Coprosma acerosa, × 6.

Fig. 73. Leaf of wild seedling Aristotelia fruticosa, changed by conditions of cultivation, × 1.