Go to National Library of New Zealand Te Puna Mātauranga o Aotearoa
Volume 44, 1911
This text is also available in PDF
(7 MB) Opens in new window

Art. I.—Observations concerning Evolution, derived from Ecological Studies in New Zealand.

[Read before the Philosophical Institute of Canterbury, 2nd August, 1911.]

Plates I-VIII.
Table of Contents.



Elementary species.










Fixity of species—plasticity.


Response to ecological factors.












After-effect of stimuli.


Convergent epharmony.


The divaricating shrub form.


The cushion form.


The liane form.


The prostrate form.


Persistent juvenile forms.




The struggle for existence.


Distribution of species.


Distribution in general.




Evolution in the genus Veronica.


Concluding remarks.


List of literature cited.

I. Introduction.

Plant-Ecology is concerned with the study of plants as living organisms, not in the laboratory under artificial conditions, but in the field as they grow naturally. Like every branch of a great science, its content is not bounded by any definite limits, but it intergrades with various departments

– 2 –

of botany, especially physiology and floristic botany, though its methods are different from those of the latter.

The conditions which the earth offers, in its manifold soils and climates, for plant-life are extremely diverse and complex, but nevertheless there exists in no few instances an apparent harmony between the conditions and the plants, which is manifested in the latter by some special form either of the organism as a whole or of one or more of its organs. It is obvious that in attempting to correlate plant-forms with their environmental factors matters are being dealt with which deeply affect the study of descent, and data are accumulated which cannot be neglected by students of general evolution.

But besides being occupied by plant-adaptations* the ecologist has also to do with the species of the taxonomist, since for one part of his work, at any rate, the groups of individuals indicated by the specific names are at present the units with which he has to deal. Furthermore, his practical acquaintance with such species, and particularly with their varieties, must in course of time become wide, while a variation with him is not merely a taxonomic mark to be noted for purposes of classification, but a physiological expression to be explained.

Besides being concerned with the origin of adaptations and species, plant-ecology deals with the arrangement of the latter into the various more or less well-defined combinations entitled “plant-associations,” and here come in such fundamental evolutionary concepts as distribution, isolation, and the struggle for existence.

Plant-ecology itself, although studied in a more or less desultory and incoherent fashion since the time of Linnaeus, may be said to date, as a special branch of botany, from the publication of Warming's Plantesamfund in 1895. At first looked at askance by the older botanists, it has steadily advanced in importance. It is prosecuted by careful and enthusiastic workers in many lands, and is now almost universally recognized as a field of the highest biological moment. Unfortunately, its methods are for the most part extremely crude, there is but little uniformity of procedure amongst its adherents, and its nomenclature is altogether unfixed. Lastly, many of the problems that await solution are amongst the most difficult that science has to offer.

Bearing the above statements in mind, it is obvious that the simpler the conditions and the fewer the species involved, the easier is it to draw conclusions of moment, and to state the ecological “facts,” if one may so designate what arise from observations made under conditions far from stringent. Also, a virgin vegetation alone can give definite information on many topics. The New Zealand biological region supplies in some measure the above desiderata. Its vascular flora, consisting of some 1,650 species, is not too great for an ecological worker to grasp; its vegetation is still in many places absolutely virgin; its climate varies from subtropical to subantarctic; some parts experience an annual rainfall of more than

[Footnote] * The convenient term “adaptation” is used throughout this paper in a non-teleological sense.

[Footnote] † This statement applies rather to the ecology of plant-distribution than to that general and more fundamental study of life-reactions known as “biology” by German investigators. In this latter sense Darwin himself stands pre-eminent as an ecologist.

[Footnote] ‡ The subantarctic and the subarctic climates are by no means identical. Intense cold plays no part in the first-named, its main characteristics being lack of sunshine, frequent cold gales, constant showers, and a low average temperature all the year, with but little frost in winter.

– 3 –

500 cm. and other parts less than 30 cm.; the plant formations include mangrove swamp, rain forest, heaths of various kinds, subglacial fell- and herb-fields, varied associations of rock and debris, subantarctic southern-beech forest, associations in and near hot springs, dunes, salt meadows, steppes, swamps, and moors—in fact, for an equal variety an ecologist would have to explore one of the larger continents in its entirety. Further, the isolation of the region for a vast period of time far from any other land-surface; the absence of grazing animals, the moa (Dinornis) excepted; the diverse floral elements (Malayan, Australian, Subantarctic, &c.); the strong endemism; the numerous small islands where conditions are simpler than on the larger ones; and, finally, the presence of many areas whose vegetation has been changed within a very few years through the farming operations of the settler, and its components replaced by exotics of quite different growth-forms—all these attributes much enhance the importance of New Zealand for ecological research.

Now, although I well know that the final court of appeal in evolutionary matters is experiment, still it seems to me that some few details having a bearing on various phases of the evolution question selected from numerous observations on a vegetation and a flora that one may venture to designate “unique” may perhaps be worth the attention of students of descent.

II. Elementary Species.

Few will deny, whatever be their opinions as to its truth, that the most awakening contribution of late years to the evolution question has been the mutation theory of De Vries. Leaving out of consideration for the present the value of the theory as a means of evolution, the introduction of careful experimental methods—i.e., a return to Darwin's own procedure—rather than mere argument in favour of this or that dogma has given new life to the study of evolution. Moreover, a change of the highest moment is the substitution of elementary species* as the raw material for the evolutionary process, rather than the Linnean species, which, as shown below, are frequently ideas merely and not living entities. It seems well, then, first of all to examine how far the doctrine of elementary species is supported by the New Zealand flora, as interpreted by ecology.

It need hardly be pointed out that the species of New Zealand taxonomists belong to the Linnean category, and that, while some refer to definite and well-defined groups the individuals of which can be recognized at a glance (e.g., Veronica Gilliesiana T. Kirk, Senecio cassinioides Hook. f., Carmichaelia gracilis J. B. Armstg., Urtica ferox Forst. f.), others vary to such an extent that there is no special set of individuals reproducing a plant that matches the specific description, which is drawn up so as to include a varying series of forms which are considered to intergrade (e.g., Veronica salicifolia Forst. f., Celmisia coriacea Hook. f., Asplenium bulbiferum Forst. f., Danthonia semiannularis R. Br., and, roughly speaking, perhaps 25 per cent of the vascular flora). Such “species” as these latter do not really exist; they are ideas only, and their origin has nothing to do with evolution. Other “species,” again, through want of a full knowledge of their

[Footnote] * This is not very different, after all, from Darwin's view, who declared that “a well-marked variety may therefore be considered an incipient species … the term ‘species’ is one arbitrarily given to a set of individuals closely resembling each other, and that it does not essentially differ from the term ‘variety.’ “(Darwin, 1899, p. 39.)

[Footnote] † And then accepting this as a species, it is said to be “extremely variable.”

– 4 –

forms, &c., may include even more than one Linnean species, as appears to be the case with Pittosporum rigidum Hook. f., as defined by Hooker, Kirk, and Cheeseman. Plate I shows this case clearly, where the type of P. rigidum on the right differs most markedly from the common South Island form on the left, which I am naming P. divaricatum.*

In some cases the difficulty as to distinguishing-names is met by the “creation” of “varieties”; but these, again, are of quite different values, and may belong to distinct biological categories. A few examples taken from the “Manual of the New Zealand Flora” (Cheeseman, 1906) will explain my meaning.

1. Hoheria populnea A. Cunn. (p. 78) is divided into the three varieties—(a) vulgaris Hook. f., (b) lanceolata Hook. f., and (c) angustifolia Hook. f. There is no such plant in existence as H. populnea, for the description includes the three varieties (a), (b), and (c), each of which, however, is distinguished by a special diagnosis, the varieties (a), (b), and (c) respectively representing distinct groups of individuals which reproduce themselves true from seed.

2. Carmichaelia Enysii T. Kirk has a variety orbiculata T. Kirk (p. 111). Both the species and its variety are described. But in this case the specific description refers to one set of individuals possessing certain characters, which is C. Enysii proper, and does not include var. orbiculata, which is to be recognized through its having other characters absent in C. Enysii proper, which latter may then be termed the type.

3. Epilobium junceum Sol. has vars. cinereum Hausskn., hirtigerum Hook. f., and macrophyllum Hausskn., each of which is defined at considerable length (p. 175). But none of these names represents a biological entity, for E. junceum, to quote from Cheeseman, “is an extremely variable plant, the numerous forms of which may be grouped in the three following varieties”—i.e., as above. Further he writes, “The extreme states of the above varieties have a very distinct appearance, and might have been treated as species were they not connected by numerous intermediate forms, which make it quite impossible to draw strict lines of demarcation between them.” Here, then, the description of the species does not indicate a type, but it includes the three varieties and all the intermediate forms, while the varieties themselves are likewise not distinct entities, and belong to a different biological category to the var. orbiculata of C. Enysii.

4. Gaultheria rupestris R. Br. (p. 407) is a similar example to the last, being said to be “a highly variable plant, the numerous forms of which are best arranged under two heads”—namely, var. lanceolata Cheesem. and var. parvifolia Cheesem.

5. In certain other cases, where there are a host of intergrading forms, the most divergent are treated as separate species notwithstanding that they are connected by intermediates. An example of this is Veronica pinguifolia Hook. f. and V. Buchanani Hook. f., of which latter species Cheeseman writes (p. 527), “Larger forms approach V. pinguifolia so closely that it is difficult to draw a line of demarcation between the two species. My var. major might be referred to either.” Other examples of similar

[Footnote] * It seems possible also that P. divaricatum consists of two elementary species, found in the steppe and forest climates of the South Island respectively. See Plate VIII, and compare it with the figure in Diels (1906).

[Footnote] † Biologically some are certainly distinct entities, as, e.g., the variety macrophyllum, which Petrie has “made” into a species under the name E. erectum, and which is greatly on the increase where forest is being removed in the Waimarino locality.

Picture icon

Example of a Taxonomic Species
On right and left, adults of Pittosporum rigidum, not distinguished as varieties, in centre, juvenile form of plant on right

Picture icon

Fig. 1.—Three Forms or the ‘Species’ Veronica Buxifolia.
Fig 2.—Juvenile Coprosma Baueri
Showing early prostrate shoots and later erect ones

– 5 –

treatment are Olearia Haastii Hook. f. and O. oleifolia T. Kirk (p. 290), Ranunculus Sinclairii Hook. f. and R. gracilipes Hook. f. (p. 18), and Poa seticulmis Petrie and P. pusilla Berggr. (p. 905).

6. Veronica buxifolia Benth., as originally defined, probably referred to a quite definite set of individuals Even by Cheeseman (pp. 522, 523) the species is spoken of as a “plant,” and not as a varying series of forms. Further, the species is defined as “erect,” and but one variety is allowed. In point of fact, however, the “species” includes three distinct growth-forms, at any rate, two of which, the prostrate, and the low, erect, sparingly branched, are shown in Plate II, fig. 1. The var. odora T. Kirk (patens Cheesem.) is of the ball-like growth-form. In this example, then, a taxonomic species includes plants belonging to at least three absolutely distinct biological categories. And, in addition, it is highly probable that a dozen or more distinct true-breeding entities might easily be separated from the heterogeneous mass of individuals known as V. buxifolia.

7. Many varieties are of a quite different physiological value to others. Some, as in cases 1, 2, and 6, reproduce themselves true from seed. This I have definitely proved in a number of instances; they are, in fact, true elementary species. Others, again, are merely environmental (unfixed epharmonic)* forms, such as are dealt with further on, of which notable examples are the var. prostrata Hook. f. of Leptospermum scoparium Forst. (p. 160), the var. rhombifolius Hook. f. of Ranunculus pinguis Hook. f. (p. 12), and the var. pauperatus§ T. Kirk of Rubus cissoides A. Cunn. (p. 125). Finally, other varieties represent a series of forms regarding the stability of which nothing is known, but which are supposed, without any sufficient reason, to be unstable.

Without going into further details, it is evident that the species of New Zealand taxonomists are rather the creation of man than of Nature. In saying this I am not hypercritical. The main object of a flora is to enable a plant to be readily identified, and this, from the very nature of the case, demands a more or less artificial classification. Where such precise and copious information as to variation is given as in Cheeseman's most careful and exact work there need be no mistake, and the worker in the field knows exactly what he may expect. But, as a rule, writers on evolution have quite neglected to distinguish between taxonomic and physiological species, which latter alone are their concern.

Although breeding-experiments can alone decide as to fixity of form, ecology should tell something. If a certain set of individuals remain unchanged over wide areas, so far as their specific marks go, and under varying conditions, it may be assumed with tolerable confidence that they reproduce their like, and are therefore species, elementary or Linnean, as

[Footnote] * Such forms are called by Massart “accomodative,” in contradistinction to “adaptive”—i.e., specific and hereditary. Regarding taxonomic varieties, the same author writes, “Malheureusement on ne peut pas toujours se fier aux travaux de systématique pour distinguer les accomodations des variations proprement dites,” and he cites the example of Polygonum amphibium, with its varieties natans, terrestre, and maritimum, all of which are simply accomodative states. (1910, pp. 9, 10.)

[Footnote] † See Cockayne, 1909, p. 16.

[Footnote] ‡ See Cockayne, 1909A, p. 201.

[Footnote] § See Cockayne, 1901, pp. 293, 294.

[Footnote] ∥ O. F. Cook's remarks are worthy of consideration (1907, pp. 362, 363): “The difficulty of defining the term ‘species’ has arisen mostly from the fact that the phenomenon is a physiological one, whereas the general supposition has been that it is morphological…. For evolutionary purposes a species is a group of inter-breeding organisms; nothing more is required, nothing less will suffice.”

– 6 –

the case may be. And perhaps it is allowable to go further, and say that if several allied plants grow in close proximity in sufficient numbers, and preserve their distinguishing characters, they are probably distinct, and would come true from seed. A case of this latter class is to be seen at the lower gorge of the River Waimakariri, Canterbury Plain, where the wars. microphylla Hook. F. and prostrate T. Kirk of Sophora tetraptera J. Mill. Grow side by side, and in this case I have proved experimentally that both varieties come true from seed. So, too, with certain forms of Acaena Sanguisorbae Vahl. Growing on subalpine fell-fields.

There is no need to multiply instances such as the above; suffice it to say that both from experiment and ecological observations I am satisfied that elementary species are very numerous in the New Zealand flora, especially in certain genera—e.g., Calamagrostis, Danthonia, Poa, Festuca, Scirpus, Uncinia, Carex, Luzula,? Phormium, Ranunculus, Cardamine, Pittosporum, Rubus, Acaena, Carmichaelia, Oxalis, Coriaria, Aristotelia, Pimelea, Epilobium, Leptospermum, Anisotome, Aciphylla, Gaultheria, Dracophyllum, Gentiana, Myosotis, Veronica, Corprosma, Celmisia, Cotula, Craspedia, and Senecio. On the other hand, many species vary to a slight degree only, and are to be recognized at a glance.

III. Variation.

Apart from constant hereditary distinctions, there are “the individual differences,” as Darwin called them (1899, p. 31), or “fluctuating variations,” as they are now frequently designated. These are supposed to depend upon a reaction of the organism to a change of environment. Klebs (1910, p. 235) distinguishes two kinds the one “caused by different external conditions during the production either of sexual cells or vegetative primordia,” and the other “is the result of varying external conditions during the development of the embryo into an adult plant.” The two sets of influences cannot as yet be sharply differentiated. The following case illustrates this difficulty.

Olearia semidentata Dcne, is a moderate-sized xerophytic shrub, which is confined to the moors of the Chatham Islands. Where both the climatic and edaphic conditions appear to be of great constancy (Cockayne, 1902, p. 288). The leaves vary on different individuals in size, shape, toothing, and tomentum, and plants grow side by side which, so far as general appearance goes, might easily be taken for distinct species. Probably here the variations are germinal. But at the same time each plant has its own rooting-place∗ and its individual physiological character, so it cannot be denied but that each plant is subjected to slightly different stimuli to those experienced by any other.

A most important question is the heredity in fluctuating variations and the degree to which they can be accumulated. Darwin (1899, pp. 31, 32) considered them all-important. “These individual differences,” he writes, “are of the highest importance for us, for they are often inherited, as must be familiar to every one; and they thus afford materials for natural selection to act on and accumulate in the same manner as man

[Footnote] ∗ The importance of the rooting-places of individuals is generally neglected by plant-ecologists who define the conditions of the habitat as a whole, whereas species growing side by side may be subjected to quite different influences, as in the case of shallow-and deep-rooting species erect and prostrate, and so on.

– 7 –

accumulates in any given direction individual differences in his domesticated productions.” And further on (p. 38), “Hence I look on individual first steps towards such slight varieties as are barely thought worth recording in works on natural history.” De Vries and his followers, on the other hand, deny that a fluctuating character can be accumulated indefinitely, and affirm that, “Selection according to a constant standard reaches its results in a few generations. The experience of Van Mons and other breeders of apples shows how soon the limit of size and lusciousness may be attained. … Improvements of flowers in size and colour are usually easy and rapid in the beginning, but an impassable limit is soon reached” (De Vries, 1904, pp. 806, 807), Further (p. 18), “Fluctuations always oscillate round an average, and if removed from this for some time they show a tendency to return to it. This tendency, called ‘retrogression,’ has never been observed to fail as it should in order to free the new strain from the links with the average.” Again, “Fluctuations are not observed to produce anything quite new, and evolution, of course, is not restricted to the increase of the already existing peculiarities, but depends upon the continuous addition of new characters to the stock.” The opinion of Klebs cannot be overlooked in this matter. This famous investigator has shown in his remarkable experiments (Klebs, 1903) that variations can be artificially induced which are far beyond the limits of fluctuating variability and considerably greater than any mutations hitherto recorded.

Ecological observations can say little on a debatable topic like this, where long-conducted experiments are alone of weight. Some observations regarding vegetables which have escaped from cultivation in New Zealand are not without interest, as showing reversion to the wild state. The radish (Rhaphanus sativus L.) is a abundantly naturalized near Wellington, but the roots are no longer swollen to any extent. The parsnip (Peucedanum sativum & Hook.), probably the celebrated “Student,” which is supposed by writers on evolution to be a fixed race,∗ came up year by year in a neglected part of my garden, but in a much deteriorated form.† So, too, with “improved” pansies primroses, and polyanthuses‡ in my garden, and with Eschschollzia californica as naturalized near Cromwell, Central Otago.

In many cases fluctuating variations are very small, and appear to be neither an advantage nor the contrary to their possessor. In other cases there are variations of much greater magnitude, which ecological observations, as shown further on, prove to be distinctly dependent upon external stimuli bringing about a response within the plant which is manifested by a visible morphological or an invisible physiological change.

[Footnote] ∗ Romanes (1895, p. 125) writes, “That is to say, it has come true to seed for the last forty years.” Romanes mentions this case as an example in support of the heredity of an acquired character, but Darwin (1905, p. 229) mentions it as a case of “methodical selection.”

[Footnote] † With a species such as this it really must be nearly impossible to judge under European conditions how far a supposed “wild” plant may be really wild and not the descendant of cultivated form.

[Footnote] ‡ The leaf-like calyx of the primroses, &c., known as “Jack-in-the-green” is a remarkably persistent character.

– 8 –

IV. Mutations.*

There seems to be no doubt but that De Vriesian mutations arise from time to time. That such afford a better material for preservation by natural selection than do small fluctuating variations is obvious. Unfortunately, the number of cases of veritable mutants is small, while most have originated in cultivation. This last fact discounts the value of the mutation theory in the opinion of many. My own feeling, as an amateur gardener of many years' standing, and as one who has cultivated with his own hands several thousand species of both wild and garden plants in an antipodean garden far from the home of most, is that ordinary cultivation, without manure, has little effect in producing variations of moment. In my garden, plants reproduced themselves from seed freely and came to maturity, but beyond a number of daffodils and some, probably hybrid, dwarf phloxes (Phlox subulata L.) I remember nothing “new.”

In estimating the origin of species by mutation, nothing but experiment can prove the heredity of the new character. All that ecology can do is to note striking varieties, their frequency, their environment, the position of the individual possessing such variations with regard to normal individuals, and so on.

The following examples of what may be full or partial mutations in the New Zealand flora, indigenous and introduced, may be of interest:—

1. The white form of Myosotidium, nobile Hook.

The species is confined to the Chatham Islands, where it grows on or near the sea-shore. In the normal form the central half of the corolla is bright blue, which fades to purple, and the edges are more or less white. Mrs. Chudleigh, of Wharekauri, some years ago discovered one plant with white flowers growing wild in the north of the main island, and although she is an excellent observer, and Myosotidium has been carefully noted in its habitat by Mr. Cox and others, no more white-flowered forms have been observed. The plant in question is now fairly common in cultivation, and, I understand, comes true from seed. So, too, does the normal blue form.

[Footnote] * Something not very different to the mutation theory was propounded by J. B. Armstrong, formerly of the Christchurch Botanical Garden, in a paper dealing with the New Zealand species of Veronica in 1881, in these words: “I have been enabled to observe numerous garden-seedlings of many of the forms, and they almost invariably resemble their parents. Sometimes, however, sports appear, and when this happens there seems to be a strong tendency on the part of the sport to reproduce itself, and it appears to me that it is just in this manner that the greater number of our native forms have been produced. At some very distant date there were probably only two or three (perhaps only one) species existing within the limits of the colony; but, on account of the extreme local variations of climate and varied geological formation of the surface, certain variations occurred, and a sport so produced, being self-fertile, and having within itself all the elements required for reproduction, naturally reproduced its like until another such sport occurred, and thus the forms gradually became differentiated from the type, and by a long series of such sports one large family of Veronicas has been formed.” Then he goes on to show how similar mutations have taken places amongst species of other lands, and considers that the intermediates have been eradicated “by man or the larger animals, leaving only in most cases the more widely differentiated forms.” But in New Zealand man has done little, and very many intermediate forms have been preserved.

[Footnote] † Klebs, however, writes (1910, p. 241), “Even if it is demonstrated that he was simply dealing with the splitting-up of a hybrid, the facts adduced in no sense lose their very great value.”

[Footnote] ‡ Raising from seed is, in fact, the only satisfactory method of propagating both the type and the white-flowered form.

– 9 –

2. The white variety of Clianthus puniceus Banks & Sol.

The type has scarlet flowers. It is now very scarce as a wild plant, but grew originally on or near sea-cliffs from the East Cape district northwards, and inland at Lake Waikaremoana. The type is a most common garden-plant; it is propagated from seeds, and comes true. According to Cheeseman (1907, p. 443), from information supplied by Mr. H. Hill, the flowers of East Cape plants vary considerably in colour, size, shape, and relative proportions of the petals. At Waikaremoana the flowers are comparatively small and reddish-purple. At Tolaga and Tokomaru they are large, and the standard very broad, with a whitish stripe on each side near the base.

The white form is white throughout. It is propagated from seed, and, according to Mr. T. W. Adams, comes true.* It is very common in cultivation. As for its origin, according to Cheeseman, “a white-flowered variety is stated by the Maoris to grow on the Tiniroto cliffs.” This may or may not be the source of the garden form. Possibly C. puniceus consists of several elementary species.

3. Geranium Traversii Hook. f. var. elegans Cockayne (Geraniac.).

The normal colour of the flowers of G. Traversii is white. It grows on coastal cliffs of the Chatham Islands. The flowers of var. elegans are pink in colour, and rather larger. It comes “true” from seed. According to Captain Dorrien Smith, it is found occasionally on Chatham Island, but I only know it as a garden-plant.

4. Phormium tenax Forst., form with purplish leaves (Liliac.).

The origin of this striking plant is not known. It is very common in New Zealand gardens It appears to come very nearly, or perhaps absolutely, true from seed, and the young plants have much more brilliantly coloured leaves than the adult.

P. tenax was commonly cultivated by the Maoris, who recognized by name many distinct-looking forms. Some of these appear to reproduce themselves more or less true, while others are probably of hybrid origin.

5. Phormium Cookianum Le Jolis, form with bracts in part instead of flowers (Liliac.). (See Williams, 1904, p. 333, and pl. 25.)

The plant in question was discovered by the Right Rev. Bishop Williams growing a little above high-water mark at Blackhead. It was then in seed, and the capsules were accompanied by numerous persistent bracts. A few of the seeds were sown. One of the young plants produced an inflorescence similar to that of the parent in 1900 and 1901, but in 1902 the four scapes produced flowers and seeds in the usual way, but these in the course of the summer “began to be clothed with leaves” in their upper portions.

6. Various crimson- and pink-flowered forms of Leptospermum scoparium Forst. (Myrtac.).

At least six individuals of Leptospermum scoparium bearing crimson or deep-pink flowers without a trace of white have been found wild in various

[Footnote] * Mr. Cheeseman informs me that he also has raised the white form from seed, and that none of the plants produced flowers other than white.

[Footnote] † Fifty-seven names are given in “Phormium tenax as a Fibrous Plant” (Wellington, 1872), but it is now known that many of them are synonyms. There are extensive collections on some of the Government experimental farms, where their behaviour as to constancy, hybridization, &c., is being studied.

– 10 –

parts of New Zealand. The two best known bear the garden names of L. Chapmanii* and L. Nicholsii* respectively. Seedlings in abundance have been raised from the latter by Messrs. Nairn and Son, Christchurch, in their nursery, and every opportunity was kindly afforded me of studying their form, &c. (see Cockayne, 1907A). The colour of the original plant is repeated more or less in the seedlings, but it varies a good deal, and some flowers are white. Dark-coloured leaves, a parental character, accompany the darker flowers.

In a case recorded by Cheeseman the plant was reported by its finder, Mr. R. J. Gilberd, to come true to colour (Cheeseman, 1908, p. 275).

It is obvious that these crimson forms only appear occasionally, for they are too striking in contrast with the familiar white blossoms to be overlooked by even a casual observer. Further, the change of colour is deep-seated in the plant, since the leaves are also affected. In L. Nicholsii Hort., too, the plant is of a weeping habit, as opposed to the normal erect stature. Finally, it must be noted that the semi-mutants grew in widely separated localities, some in the South and others in the North Island.

7. Double white form of Leptospermum scoparium Forst. (Myrtac.).

This was found growing wild on pumice soil in the Hot Lakes district by Mr. E. Philipps Turner. The doubling is very complete, and, so far as I could judge from much-damaged specimens, resulted from petalody of the stamens. Probably it is unable to produce seed. This case is of further interest because double flowers, as De Vries has pointed out (1905, p. 489), are exceedingly rare in the wild state, though so common in cultivation. Only one individual was noted. The mutation was evidently quite spontaneous, and cannot be attributed to any sudden change of soil-conditions.

Leptospermum scoparium is a most variable plant. Doubtless some of the forms are good elementary species. The form with pinkish flowers and hairy leaves, &c., of northern Auckland, which occurs over wide areas side by side with other forms from which it can be recognized at a glance [ unclear: ] is a case in point.

8. Olearia semidentata Dcne., form with white florets.

The type has brilliant purple flower-heads. The white form was discovered growing wild by Captain A. A. Dorrien Smith. It is now in cultivation in the garden at Tresco Abbey, Scilly.

A similar case is var. Dendyi Cockayne of Olearia chathamica T. Kirk, found on Pitt Island by Dr. A. Dendy, F.R.S., and which has purple florets and yellower denser tomentum on the under-surface of the leaf than the type, the florets of which, moreover, are white fading to purplish.

9. Metrosideros lucida Menzies, form with white flowers.

The type has crimson flowers. The white-flowered form has been found in two places, one plant which I have seen growing near the head of the Otira Gorge, Westland, and the other lower down the valley.

10. Metrosideros tomentosa A. Rich., form with yellow flowers.

Mr. H. Carse (Cheeseman, 1906, p. 1137) discovered one specimen with yellow flowers, those of the type being dark crimson, at Rangaunu Harbour, northern Auckland.

[Footnote] * Because I use these garden names it must not be concluded that I consider the plants of the same biological class as Limnean species, or even elementary species.

[Footnote] † I am indebted to Mr. J. O'Malley, of Otira, for calling my attention to the latter plant.

– 11 –

11. Rubus Barkeri Cockayne.

This is a presumably non-flowering species (see Cockayne, 1910, p. 325); at any rate, cuttings from an adult plant growing luxuriantly and under most varied conditions for a period of thirteen years have never flowered. Be this as it may, the plant in question, although closely related to R. parvus Buchanan, differs from that species in its trifoliate leaves with lanceolate* leaflets and not simple linear leaves, its serrate and not dentate leaf-margins, its different autumnal colouring and its greater size in all parts (see fig. 1).

Picture icon

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

Fig. 1.—Outline of Leaf of (a) Rubus Barkeri, (b) R. Parvus. × 2/3

Only one plant was originally noted. Recently I have seen abundance of Rubus parvus in various localities in Westland and under different conditions, but it is remarkably constant in characters, and presented no transitions towards R. Barkeri. I know well that my action in “creating” a taxonomic species in this case is open to adverse criticism, especially as I believe that the original wild plant may be the only one in existence; but if a species can originate by mutation there must be a time when there is only one individual, and if so, and its characteristic marks are of “specific” importance, it is just as much a “species” as if there were thousands of similar individuals.

[Footnote] * So defined in original description, but leaflets in fig. 1 are broader.

– 12 –

12. Veronica Benthami Hook. f., form with white flowers.

V. Benthami is a shrub of straggling habit endemic in the Auckland Islands. The flowers are normally of a brilliant blue, a most unusual colour amongst New Zealand plants. One or two individuals with white flowers were noted by me in 1907. Also, another plant had the flowers almost carmine when just opening, but fading to a paler colour on the outer parts of the corolla when fully expanded (Cockayne, 1909, p. 203).

13. Occurrence of variegation, &c.

There are three forms of variegated Coprosma Baueri Endl. in cultivation of whose origin I know nothing. A variegated form of Griselinia littoralis Raoul was discovered a number of years ago by the late Mr. Purdie in the vicinity of Dunedin. The late Mr. H. J. Matthews found, also in the neighbourhood of Dunedin, a form of Fuchsia excorticata L. f. with very dark-coloured leaves, quite different from the normal. One individual of Cordyline australis Hook. f. with variegated leaves was found many years ago in a batch of seedlings raised at Duncan's nursery, Christchurch. It appears to come true from seed. Variegated forms of Veronica salicifolia Forst. have appeared on several occasions in cultivation. There are variegated forms of Pittosporum tenuifolium Banks & Sol. and P. eugenioides A. Cunn., but their origin is unknown. A form of Coprosma robusta Raoul with yellow and not the typical red-orange drupes was found by me near Kaipara Harbour, Auckland. There are a number of variegated forms of Phormium tenax Forst. and P. Cookianum Le Jolis in cultivation, which come more or less true from seed, but a variegated plant of the latter species found wild by me on Mount Sherwood, Marlborough, upon being brought into cultivation reverted to the type.

14. Tetragonia expansa Murr.

This case is cited by De Vries (1901, p. 469). There are two forms, one with brownish and the other with green flowers; both came true. The wild plant in New Zealand has yellow flowers.

15. Pittosporum tenuifolium Banks & Sol., form with yellow flowers.

In New Zealand, so far as is known, the petals are invariably dark-purple, almost black. But, according to H. M. Hall (1910, pp. 7, 8), two shrubs growing in a row of the normal-coloured plant in California produced yellow flowers. Should this be at all common in New Zealand it could hardly have escaped notice.

16. Introduced plants.

Some remarkable more or less hereditary variations have come about in the broom (Cytisus scoparius Link.), gorse (Ulex europaeus L.), and tree-lupin (Lupinus arboreus Sims). In the first two named species there are colour-changes from the normal yellow to white, differences in size and shape of flower, and, in the gorse, variation in time of blooming.

Lupinus arboreus Sims, normally yellow, and varying but little in its native land, on the dunes near New Brighton, Canterbury, has undergone many remarkable changes in the colour of its flowers. There is, e.g., a pure-white, yellows of various tints, and a great variety of purples combined, or not, with whites and yellows. These abnormally coloured plants occur in patches here and there as a general rule, and appear to get more abundant year by year. In the North Island I have neither noticed nor heard of such variations, nor yet in Central Otago.

– 13 –

Red clover (Trifolium pratense L.) and cowgrass (the var. perenne) vary to an astonishing extent in a small patch, chiefly self-sown, in my garden. Many of the forms are most distinct, and the new characters are diverse, affecting colour of flowers, stems, and foliage, form of inflorescence, degree and kind of hairiness, general habit, &c. Here pure culture-methods and Mendelian procedure would be needed to come to any reliable conclusions as to variants such as these.

Holcus landtus L. and Dactylis glomerata L., I am informed, vary at times beyond their ordinary fluctuating capacity.

Capsella Bursa-pastoris Medic., a very variable species in its natural habitat, and which has already given rise to certain mutants, varies to an astonishing degree in New Zealand, especially in highly manured ground. A careful study of such variation is certainly demanded.

V. Epharmony.

1. General.

It is when we come to epharmonic adaptations that ecology presents its most important contribution to the evolution question.

In attempting to explain the origin of epharmonic adaptations it is evident that, as in the case of all explanations of evolutionary phenomena, no absolute proof can be given without experiment, and, where such is wanting, it seems reasonable that the most probable explanation should be accepted for the time being, notwithstanding that other though less probable explanations would fit the case. Generally in polemical discussions on matters of evolution natural selection is assumed to be a vera causa which needs no demonstration, and if any other reason is put forward, however likely it may appear, it is considered of no moment, unless it can be proved not merely to the hilt, but to the objector's satisfaction.

Now, I am of opinion that in the hereditary epharmonic variations cited below there is a much greater likelihood of their having been brought about by the direct action of the various ecological factors than by the continuous accumulative selection of fluctuating varieties, and in making this statement I am merely echoing the opinion regarding analogous phenomena of Romanes (1895, pp. 122–32), Warming (1909, pp. 370–73), MacDougal (1911, p. 57), Henslow (1895, 1908), Costantin (1898), (Scott-Elliott, 1910), and many other writers on evolution.

With the much-disputed Lamarckian factors use and disuse, which are so frequently the only parts of the doctrine dealt with by the zoological opponents of modified Lamarckism, I have nothing to do. How far evolutionary methods correspond in the plant and animal kingdoms no one can say, but it does not seem unreasonable to imagine that they may have been in many respects different.* At any rate, this paper is concerned only with the botanical side of evolution.

2. Fixity of Species—Plasticity.

Nothing has been brought out more clearly by ecological studies in New Zealand than the extreme “plasticity” of many species and structures, and their rapid response to a change of environment. This is so

[Footnote] * Leavitt (1907, p. 237) writes, “In no case is it safe to reason deductively from one kingdom to the other. In the factors affecting their evolution, plants and animals differ vastly.

– 14 –

great in numerous instances that the idea of “normal” loses its meaning. Take the following examples:—

(a.) Leptospermum scoparium Forst. (Myrtac.) may be a moderate-sized tree, a tall shrub, a dwarf plant 2–8 cm. tall which flowers and ripens seed, and an absolutely prostrate plant which forms a dense covering to the ground and puts forth adventitious roots, although the erect forms are exceedingly difficult to artificially strike as cuttings.

(b.) Certain shrubs are of the xerophytic divaricating growth-form when growing in the open, but of a comparatively loose, leafy, and mesophytic habit when growing in the shade and shelter of the forest—e.g., Pittosporum divaricatum* Cockayne, Corokia Cotoneaster Raoul, Aristotelia fruticosa Hook, f., &c. In such a case, were the shade form alone in existence (see Plate VII, fig. I), there is no botanist but would consider it fixed and normal, and yet it is the sun and wind form rather that is so considered.

(c.) Fuchsia Colensoi Hook. f. (Onagrac.) is a twiggy shrub in the open, but in the forest it is frequently a scrambling liane.

(d.) Hymenophyllum multifidum Sw. (Filic.) when occupying wet rocks in the Auckland Islands has its fronds closely curled up, but when growing in the forest interior of the same group they are quite flat. That the curled fronds are not fixed I have shown by means of moist-air culture (1904, pp. 266, 267). Suppose that H. multifidum was only found on a windswept treeless island, such as Macquarie Island, no one would question the curled frond being normal and fixed.

(e.) Myoporum laetum Forst. f. (Myoporac.) is nearly always a small round-headed coastal tree having a distinct erect trunk, but on Moko Hinou Island it is altogether prostrate, and its branches far-spreading, cord-like, and twiggy. Were it not that I have seen intermediate forms on some parts of the North Island coast I could hardly believe that the Moko Hinou plant was not a stable form.

(f.) Myrtus pedunculata Hook. f. (Myrtac.) is generally either a small tree or a twiggy erect shrub, but at an altitude of some 1,200 m. in the Nothofagus forest of the volcanic plateau, North Island, it is frequently quite prostrate and rooting. Styphelia fasciculata Diels (Epacrid.), although nearly always an erect shrub as a forest-plant, behaves exactly as the last-named in the same station. On dunes it is also frequently prostrate.

(g.) Dracophyllum politum Cockayne (Epacrid.) when growing on the mountains of Stewart Island is a turf-making shrub, a low spreading shrub with stout horizontal branches, or a massive ball-like cushion plant, according to circumstances. So different are these various forms that I can hardly yet believe them to be merely environmental unfixed forms of one another§ and that my observation is not faulty.

(h.) Gleichenia dicarpa R. Br. and G. circinata Sw. (Filic.) differ specifically in the former having the margins of the segments of the pinnae incurved so as to be pouch-shaped, whereas those of the latter are virtually flat. But the same individual of the var. hecistophylla Hook. f. will possess

[Footnote] * This plant has been merged with P. rigidum Hook. f. A diagnosis has not yet been published, owing to lack of sufficient material, but it is necessary here to use a definite name, since P. rigidum and P. divaricatum are certainly distinct entities. (See Plate I.)

[Footnote] † For further details, see Cockayne, 1901, pp. 265–67, and Diels, 1906, pp. 66–69.

[Footnote] ‡ It is possible that the tree and shrub are different species, but I hardly think so, though I have not seen intermediates.

[Footnote] § See Cockayne, 1909, p. 16, and photo No. 13, facing p. 17.

– 15 –

some pinnae with pouches and others quite flat, in accordance with the degree of illumination to which they are exposed. In fact, here the specific distinction does not hold—it is merely epharmonic—and the latest name must be abandoned; nor can the two “species” be maintained even as “varieties.”

(i.) Discaria toumatou Raoul (Rhamnac.) when growing in positions subject to the attacks of rabbits may form low green cushions made up of leafy spineless shoots. “Normally” it is a stiff branching shrub furnished with abundant spines.

Many more examples could be cited, but the above show clearly enough how unstable species may be, even when growing under natural conditions. When experimental methods are brought into play the effects from plasticity become still more striking. For example, spine-production may be suppressed in Discaria toumatou; true leaves may be produced in the whipcord veronicas and species of Carmichaelia (Legum.); rolled leaves made flat,* and vice versa; cushion plants opened out widely. Undoubtedly a series of experiments such as those of Klebs (1903) would yield results equally surprising.

It can be seen from the above that this uncertainty as to “normal” form opens up room for great doubt in all discussions regarding the origin of permanent adaptations, for it may quite well be asserted that absolute fixity does not exist. It seems to me all that can be done is to consider as “normal” those forms which predominate and represent the general growth-form of the bulk of the individuals; but assuredly in no few cases there is no normal form at all.

3. Response to Ecological Factors.

Warming has summed up the state of knowledge on this head up to the date of publication of his admirable “Oecology of Plants” (Warming, 1909, pp. 16–81), so that only a few local examples are necessary here. First of all, it must be emphatically pointed out that it is virtually impossible in the field, where so many ecological factors are concerned, to say which is the predisposing cause of the internal response of the plant. Generally more than one factor will be concerned.

(a.) Soil.

Excess of salt leads to succulence, as in certain salt-meadow species which become less succulent as members of non-halophytic formations. The introduced Silene anglica L. develops more succulent leaves when growing near the sea than inland. Miss Cross examined the anatomy of certain salt-meadow plants and those of the same species grown in ordinary soil in a greenhouse. Her figures show considerable differences in thickness of leaves, but other factors besides want of salt doubtless affected the result (1910, pp. 569–71).

The soil near hot springs containing excess of sulphur, &c., inhibits the erect shrub form of Leptospermum ericoides A. Rich., which then occurs only in the prostrate form.

Lack of nutritive salts in sand-plains near the mouth of the River Rangitikei and elsewhere changes the leaf-form of Selliera radicans Cav. (Gooden). This is in accordance with the much more carefully conducted observations

[Footnote] * In the case of Olearia cymbifolia Hook. f. the much revolute, boat-shaped leave become flat with moist-air culture, and what was considered an important specifi character, distinguishing the “species” from O. nummularifolia Hook. f., vanishes.

– 16 –

of Massart, which are supported by soil-analyses (1910, pp. 156–65). The prostrate habit of certain shrubs of dune-hollows in the north of Auckland may, in part, be similarly explained.

Acid peat soils favour the cushion and other xeromorphic growth-forms, though mesophytic forms may also occur.*

Phyllachne clavigera F. Muell. (Stylid.), and doubtless its allies of similar cushion-form, can be made of much looser growth by moist-air culture (Cockayne 1909A, p. 201).

The shoots of Cotula Haastii T. Kirk (Compos.), one part of a plant rooted in deep soil, and another part on rotten rock or shallow soil, exhibit certain striking differences. These are chiefly in degree of intensity of characters. The portion in shallow soil has smaller leaves, stiffer stems, more glands, and the leaf-segments closer. The leaves are of a darker green, and are marked with brown on the lower half, whereas there is no trace of brown on the deep-rooting portion. A dune form of Acaena microphylla Hook. f. behaves similarly in my garden, the leaves of non-rooting shoots being much smaller than those of rooted shoots and broadly margined with brown, the “normal” leaves being lighter green and faintly brown at the apices of the teeth at most. This presence or absence of a dark colouring-matter would appear of small importance were it not that dark-coloured leaves are a rather frequent characteristic of New Zealand plants.

Plants exposed to drifting sand may develop an upward growth. Thus, Poa caespitosa Forst. f., although a steppe tussock-grass, when growing on drifting sand in Central Otago gets more or less a sand-binding form. So, too, with Phormium tenax Forst. and Arundo conspicua Forst. f. on coastal dunes, though both are commonly swamp-plants.

Scirpus frondosus Banks & Sol., a sand-binding plant of the most extreme type, is not only endemic, but belongs to an endemic subgenus (Desmoschoenus). Not only has this plant attained its growth-form in an isolated dune-area, but, as Mr. R. B. Oliver suggests in a letter to me, possibly in actual competition with the Australian Spinifex hirsutus Labill.

At one place in Puhipuhi Valley, Seaward Kaikoura Mountains, nearly all the species, both indigenous and introduced, growing on cold, wet, limestone soil exhibit marked variegation, but beyond this edaphic influence they are of the normal green.

Highly manured soil, as is well known in cultivation, acts powerfully upon plant-form. In nature the same occurs. Plants of Sicyos australis Endl. growing on ground manured by Puffinus sphenurus in the Kermadec Islands frequently produce male flowers in which “the petals turn green, and assume more or less the shape and character of foliage leaves” (Oliver, R. B., 1910, p. 132). Certain species appear confined to soil of the above character—e.g., Senecio antipodus T. Kirk, of Antipodes Island and Cotula Featherstonii F. Muell., of Chatham Island.

(b.) Light.

The bright light of dunes probably leads to the red- or orange-coloured stems of the rush-like Leptocarpus simplex A. Rich (Restiac.), which are green in the shade, and as salt-swamp plants not nearly so brilliantly coloured. It is a moot point how far the reddish, yellowish, or brownish hue of certain true dune-plants may be considered fixed and hereditary

[Footnote] * See on this head Burns, 1911, pp. 121,124. Xerophytes are confined to certain zones in the bogs studied, the largest bog-areas being hydrophytic or mesophytic.

– 17 –

(e.g., Scirpus frondosus Banks & Sol., Coprosma acerosa A. Cunn., Gunnera arenaria Cheesem., Euphorbia glauca Forst. f.).

An interesting case is that of Lycopodium ramulosum T. Kirk, a plant forming extensive patches on moors in the west of the South Island and Stewart Island, the sporophylls of which are absent or scantily produced in shade plants, but extremely abundant in those growing in bright light.

Many young trees in the forest assume a special form with a slender main stem and few branches, which are confined to its upper portion. Similarly, the xerophytic fern Pteridium esculentum Cockayne becomes in a dim light a scrambling liane. An example observed by Mr. H. Carse and myself was growing amongst tall, slender Leptospermum scoparium on Reef Point, north-west Auckland. Some of the fronds were more than 3 m. in length. Pinnae were absent until the brighter light was gained. The final portion of the rhachis was green and succulent, and the distance between the pinnae 46 cm. These latter were still coiled up and quite rudimentary, although the largest was 25 cm. long. The rhachis was twisted—i.e., it showed a tendency to twine.

Shade—and here probably comes in moisture in the air—increases the size of leaves, changes certain xerophytes into mesophytes: e.g., species of Carmichaelia, Discaria toumatou. Podocarpus nivalis Hook., as may be plainly seen from Plate IV, responds markedly to changes in illumination, the shade form resembling P. totara much more than the species which it really is. The specimens were collected within a few feet of one another.

The lie of the leaf is regulated by the light. Olearia insignis Hook. f., a shrub of dry rocks in Marlborough, arches its branches upwards to a surprising degree, thus bringing its leaf rosettes into a suitable position with regard to the light. This habit persists in plants raised from seed and grown on flat ground.

(c.) Wind.

Wind is a most important factor in New Zealand. First comes the “wind-shearing” action, which is in part a physiological process; it is well marked in trees and shrubs of exposed positions, and may be frequently seen in Podocarpus totara D. Don., Leptospermum scoparium Forst., and many other plants. The prostrate habit is encouraged by wind; but here other factors may enter in, as cold and acid soil. Coprosma foetidissima Forst. is usually a tall forest-shrub, but when a member of the tussock-moor association of the Auckland Islands (Cockayne, 1909A, pp. 200, 201, and 219) it is prostrate and twiggy. The prostrate form of Leptospermum scoparium on the subalpine moors of Stewart Island is another and remarkable example. Well-developed prostrate trunks are to be seen in Metrosideros lucida Menz. (Myrtac.) in the Auckland Islands, Stewart Island, and the Southern Alps, and in Olearia ilicifolia Hook. f. (Compos.) in some subalpine forests of the South Island. Reduction in size of leaves must often be attributed to wind-action.

(d.) Water.

Plants of still or slowly running water are subjected to a fairly constant environment.* Cotula coronopifolia L., as a land-plant, is a herb with branched, prostrate, more or less rooting stems, the branches of which are erect or semi-erect; the internodes are short; the leaves are rather fleshy, more or less lanceolate in outline, and pinnatifid, lobed, toothed, or sometimes entire; the roots are, at most, of a moderate length. As a

[Footnote] * Of course, the position of the plant with regard to the surface, the nature of the substratum, and other factors exercise a considerable influence.

– 18 –

water-plant, the stem is straight, unbranched, and perhaps 40 cm. long; the internodes are long; the leaves linear and entire, and the roots numerous and 30–40 cm. long; when the shoot rises above the water-surface it branches, and the leaves are much as in the land-plant.

Not only the leaves but also the inflorescence differ greatly in size in the land and water forms of the introduced Radicula Nasturtium-aquaticum Brit. & Rend.

Specially moist air causes the production of aereal roots on the stems of certain whipcord veronicas.

Schefflera digitata Forst. (Araliac.), a low forest tree or shrub, when growing in certain damp gullies of northern Auckland produces sometimes leaves much more deeply cut than the normal.

The moist-gully form and the dry- or acid-ground form of Blechnum capense Schlcht. (Filic.) are so distinct in appearance that many might consider them distinct species.

(e.) Altitude.

Altitude is a complex combination of factors which sometimes produces striking differences in the same species, according to the height at which the individuals grow.

A very common feature is diminution of stature with increase of altitude, though this is not so with all species. The trees Dacrydium cupressinum Sol., Weinmannia racemosa Linn. f., and Griselinia littoralis Raoul are much reduced in size when forming a part of the mountain-scrub of Stewart Island, the two latter eventually becoming small shrubs.

On the other hand, if the lowlands can offer an equivalent environment to that of the mountains—though, of course, it can never be actually identical—alpine plants may occur at sea-level, their forms differing not at all from those at an altitude of 600 m., 900 m., or considerably higher. The lowland moor of Stewart Island contains various alpine plants of this character—e.g., Celmisia argentea T. Kirk, Astelia linearis Hook. f., Dracophyllum politum Cockayne, Carpha alpina R. Br., Donatia novaezelandiae Hook, f., Caltha novae-zelandiae Hook. f., Gaimardia ciliata Hook. f. (For full list, see Cockayne, 1909, p. 27.)

4. After-effect of Stimuli.

It is most important with regard to the question of the ultimate heredity of changes in form and structure, &c., brought about by an internal response of the plant to stimuli from without to inquire as to definite examples where the form, &c., persists for a reasonably long time after the stimulus is removed. The following cases bear on this subject:—

A prostrate form of a species of Coprosma (Rubiac.), which originally grew on acid peat on the Chatham Island tableland, was cultivated by me in a pot for three years, and then in ordinary garden-soil in a garden for four years more, during the whole of which time the prostrate habit remained. But all on a sudden, during the eighth year, it commenced to put forth erect shoots, and but for its unfortunate destruction would undoubtedly by this time have been on erect shrub. So assured was I that this plant would remain prostrate or stunted that I published certain remarks to that effect (1907, p. 378). So, too, with a stunted form of another species of Coprosma, perhaps C. cuneata Hook. f., collected by me in 1903 in Antipodes Island. This was grown on the rockery at

– 19 –

Canterbury College for six years and kept its habit, but later on it too commenced to put forth erect shoots.

Coprosma Baueri Endl. when growing on a sea-cliff is a straggling shrub, more or less closely flattened to the rock-surface, and puts forth nothing but long spreading horizontal shoots. Such plants bear flowers and fruit. This growth-form of the species may be referred to wind, and perhaps heat. But when C. Baueri grows in a coastal forest, or even when isolated on loamy clay, it is a tree with a stout trunk. Plants which I raised from seed, and which are now growing in the experiment-ground at Canterbury College, possess long spreading horizontal shoots—i.e., they are of the shrub form, as above; but they are also developing erect shoots, and, if permitted, they will eventually grow into trees (see Plate II, fig 2). Here it is possible that the prostrate form is inherited from the race of rock-frequenting plants. But the stimulus has not been sufficient to make a really permanent race, and so the prostrate form only occurs during an early stage in the ontogeny of the individual. Similar cases of partial heredity are dealt with further on when treating of prolonged juvenile forms.

Olearia Lyallii Hook. f. (Compos.) forms a pure forest on some of the New Zealand subantarctic islands. A striking feature is the prostrate or semi-prostrate trunk, which may be referred to wind, a peat soil, and perhaps a uniform low temperature. In the interior of the forest, no matter how boisterous is the wind without, it is quite calm, and yet the seedlings are nearly always more or less prostrate at first. So, too, with the seedlings of O. Colensoi Hook. f. when growing on the mountains of Stewart Island.

The case of Sophora microphylla Ait. and S. prostrata Buchanan: This is fully discussed in this paper under the heading “Persistent Juvenile Forms” (p. 25), to which it may be well perhaps for the reader to turn and consider the case in relation to the point under discussion.

It would be beyond the scope of this paper to mention in detail instances of after-effect of stimuli in places other than New Zealand, but it is well to briefly enumerate a few of the more striking. Such are Schübler's cereals, which, grown in a northern climate, ripened their seeds earlier even when cultivated in southern countries; Cieslar's conifers, whose seeds, collected in the Alps, when sown on the plains produced plants of slow growth and small diameter; Klebs's Veronica and Sempervivum, whose striking abnormalities of inflorescence were repeated in plants raised from seed; Blaringhem's races of maize and barley originating from plants purposely damaged in a specific manner (Blaringhem, 1907); Zederbauer's experience with a form of Capsella Bursa-pastoris from an altitude of 2,000–2,400 m. in Asia Minor, which through four generations in Vienna maintained in part the special alpine stamp; and MacDougal's ovarial treatments, where one new induced form has maintained its character, so far, up to the fifth generation (see MacDougal, 1911, pp. 56, 57).

5. Convergent Epharmony.

From what has gone before, it is plain that various growth-forms of New Zealand plants may be referred with confidence to stimuli from outer factors. It has been seen also that of such forms some are merely environmental; but there are others, now to be dealt with, which are hereditary, and remain constant, unless perhaps when exposed to such a change of conditions as they would not encounter in nature.

– 20 –

It is a fact of the greatest significance that identical growth-forms are found side by side amongst species belonging to unrelated families. The importance of this occurrence is still more emphasized by the fact that other species in far-distant parts of the earth, growing under approximately similar conditions, may likewise possess the same epharmonic forms. That there should be this convergent epharmony, as it is called, seems to lend the strongest support to the view that the effect of an outer stimulus upon the plant, such as light, heat, &c., may become hereditary.

Only a few characteristic growth-forms receive attention here, and the treatment of these is quite brief. A few others are dealt with when treating of the genus Veronica (p. 44).

(a.) The Divaricating Shrub Form.

This very common New Zealand growth-form consists of much-branched often stiff and wiry stems which are pressed closely together or even interlaced, the branching being frequently at almost a right angle (see Plate III, fig. 2). Although I do not know of any example where wind has brought an exact replica of this form, a wind-shorn shrub is closely related. Still more close is the unstable form assumed by certain lianes in the open (e.g., Rubus, Muehlenbeckia,* and Clematis) which grow in company with true divaricating shrubs. Further, the relation to shrubs of an open growth is exhibited by the already mentioned Corokia Cotoneaster and Pittosporum divaricatum, when they grow as forest-plants. Suttonia divaricata Hook. f. (Myrsinac.) is virtually fixed under all circumstances, though in the forest it may have a slender trunk.

The ecological factors governing this growth-form appear to be wind, in the first place, and then various other xerophytic stimuli, of which soil must play an important part.

The most instructive case of convergent epharmony in these plants is in the scrub of certain South Island montane river-terraces or riverbeds, where so greatly do many of the species resemble one another that it is quite easy to confuse them. The following is an actual combination: Pittosporum divaricatum Cockayne (Pittosp.), Rubus subpauperatus Cockayne (Rosac.), Discaria toumatou Raoul (Rhamnac.), Aristotelia fructicosa Hook. f. (Elaeocarp.), Hymenanthera dentata R. Br. var. alpina, T. Kirk (Violac.), Corokia Cotoneaster Raoul (Cornac.), Coprosma propinqua A. Cunn., C. parviflora Hook. f. (Rubiac.). Hymenanthera would frequently be absent or confined to specially stony ground. There would also probably be one or more species of Veronica and Carmichaelia, but their growth-forms are different.

The divaricating growth-form also occurs in the following families: Polygonaceae, Ranunculaceae, Leguminosae, Rutaceae, Icacinaceae, Malvaceae, Mysinaceae, and Compositae—i.e., in fifteen families altogether, all of which have likewise members with altogether different growth-forms. Generally speaking, the earlier juvenile form of these plants is mesophytic.

(b.) The Cushion Form.

Every transition exists between the open circular mat-like form and dense unyielding cushions. It is merely a question of degree in reduction of internodes and closeness of growth. The genus Celmisia shows

[Footnote] * M. Astoni Petrie, most closely related to the liane, M. complexa, is a divaricating shrub.

Picture icon

Fig. 1 —Sophora Microphylla
Juvenile divaricating form
Fig. 2.—Pittosporum Divaricatum
A shrub of the divaricating growth-form

Picture icon

Podocarpus Nivalis
On left, shade form; on right, sun form Plants from Otira Gorge

– 21 –

straggling mats in C. discolor Hook. f. and C. Walkeri T. Kirk, loose circular cushions in C. viscosa Hook. f., and true dense cushions in C. sessiliflora Hook. f. and C. argentea T. Kirk.

Frequently the epharmony of such cushions can be seen clearly in one and the same species, as in the tiny taxad Dacrydium laxifolium Hook. f., which forms cushions on dry pumice at 1,200 m. altitude near Mount Ruapehu, but which growing amongst other shrubs under more mesophytic conditions is frequently a straggling shrub, or when in colonies on sour peaty ground merely a close turf.

The cushion form culminates in the great amorphous masses of certain species of Psychrophyton and Haastia, which grow on alpine rocks* exposed to sun, frost, and wind, or at times, in the case of R. Goyeni T. Kirk, of Stewart Island, on wet peat.

Excepting with regard to the physiologically different bryophyte cushions of moors or wet forests, the cushion form is governed by strong xerophytic conditions, and the same species may thrive either in physically or physiologically dry stations—e.g., Phyllachne Colensoi Berggren (Stylid.), Psychrophyton Goyeni Beauverd (Compos.).

The form under consideration occurs in the following families: Taxaceae, Gramineae, Cyperaceae, Centrolepidaceae, Juncaceae, Portulacaceae, Caryophyllaceae, Leguminosae, Violaceae, Thymelaeaceae, Umbelliferae, Boraginaceae, Scrophularinaceae, Plantaginaceae, Stylidiaceae, and Compositae.

Epharmonically similar cushions occur amongst different genera and families in high mountains everywhere. Certain erect shrubs when windswept become virtually cushions.

(c.) Lianes.

Climbing-plants have most certainly descended from non-climbing species which through shade and moisture have grown upwards out of the lower tiers of vegetation in a stratified association. Many transitions between climbing and non-climbing plants can be observed, and these, considered along with the heredity of the climbing habit and its strong differentiation, afford weighty support to a belief in the heredity of epharmonic characters.

The fern Hypolepis distans Hook., which generally gives no hint of a propensity to climb, when growing alongside a support may lengthen its fronds for considerably more than I m., though at this length they would fall but for the support. On the rhachis are minute excrescences, which, though certainly not adaptations for the purpose, assist the frond to maintain its position. The climbing form of Pteridium esculentum, already noted, is specially interesting because of its hint at winding. So, too, with the scrambling liane Lycopodium volubile Forst. f., which, gaining a thin support, winds freely, the winding being in this case an hereditary characteristic.

The case of Fuchsia Colensoi Hook. f., already mentioned, is of especial moment. This is a shrub in the open, and at times a scrambling liane in the forest. There can be little doubt that this latter habit is hereditary to some extent, and it is possible that there may be climbing and non-climbing races. This is the more likely as the “species” is considered variable, and large forms are said to “almost pass into F. excorticata” (Cheeseman, 1906, p. 187), which is a small tree or shrub, but never a liane.

[Footnote] * Haastia pulvinaris appears to grow on shingle-slip, and not on rock, so far as I have observed; but I am also advised that at times it grows on rock.

[Footnote] † Strictly speaking, there is no “purpose” in any adaptations, but it is often convenient to speak teleologically.

– 22 –

In the case of Rubus cissoides A. Cunn. var. pauperatus T. Kirk there is no question of distinct races, although there are certainly two epharmonic growth-forms. The one is a high-climbing liane growing in forests. It is provided more or less abundantly with leaves, and produces plenty of flowers and fruit. But in the open, on hillsides fully exposed to wind and sun, it forms rounded bushes of interlacing twigs, has its leaves reduced to midribs, and never produces flowers. It is, in fact, a xerophytic form, governed by the non-forest conditions, and its presence depends upon seeds being brought from the forest-plant by birds. Seedlings raised by me from the forest-plant were leafy in an early seedling stage*; this was followed by the epharmonic leafless form, which, although hereditary, can only persist so long as xerophytic conditions are maintained. Plants growing in the shelter of a cliff may have a few leafy shoots which can bear flowers and fruit. Rubus subpauperatus Cockayne, closely related both as a species and as a growth-form, has also a forest form and a xerophytic form, but in this case both produce flowers. The adult flowering forms of the root-climbing lianes Metrosideros scandens Sol. and M. florida Sm. may become shrubs in the open, an analogous case to that of the artificially raised tree-ivy of gardens. It is highly probable that other climbing species of the genus behave in a similar manner.

The genus Clematis is represented by eight species in New Zealand. All are more or less variable, and some of the species appear to “run into” one another. Six may be considered mesophytes; they are forest-plants, or some climb amongst shrubs. These species are abundantly furnished with leaves. But the var. rutaefolia Hook. f. of C. Colensoi Hook. f. grows under more xerophytic conditions, and, in accordance with these, it is smaller than the type, the leaves are more cut and present less transpiring surface; perhaps it is a fixed form. C. marata is subxerophytic; it grows in the open, frequently climbing into the branches of the xerophytic Discaria toumatou; its stems are slender, brownish-green, pubescent, and interlaced, and its leaves much reduced. Finally, C. afoliata Buchanan is a true xerophyte; it is virtually leafless; the stems are green and function as leaves; they are rush-like, grooved, have the stomata in the grooves, and are generally closely intertwined — i.e., the growth-form is identical with that of the above Rubus, and approximates to the divaricating form. The seedling has plenty of leaves, and when the adult grows in the forest this juvenile state may persist and even flower. It must be remembered that this range of forms of Clematis, which vary from forest mesophytes to an almost divaricating leafless shrub form, are all presumably descended from one ancestor, and that even now many are connected by intermediates, while one species is epharmonically mesophytic or xerophytic, according to its station.

(d.) The Prostrate Form.

There are various modifications of the prostrate form, which depend chiefly upon closeness of branching and rooting-capacity. Here there are only mentioned those with more or less straggling stems, which may or may not bear adventitious roots. On certain subalpine moors a number

[Footnote] * See also figs. 229, 230, Goobel, 1905, pp. 353, 354.

[Footnote] † C. quadribracteolata Col. is omitted, as it seems to me merely a variant of C. marata J. B. Armstrong. Nor do I know anything regarding the vars. depauperata Hook. f. and trilobata Kirk of C. parviflora A. Cunn.

Picture icon

Fig 1 — Veronica Chathamica.
Cutting planted vertically but growing horizontally. thus showing an hereditary character
Fig 2
1 Veronica Loganioides 2 V Cassinioides 3 Reversion Shoot Of V Tetragona. 4 V Tetragona × 3

Picture icon

Fig 1 — Sophora Tetraptera
Young tree of Chatham Island form growing eject with straight branches
Fig 2 — Sophora Tetraptera
Seedlings of Chatham Island form

– 23 –

of plants of this class may grow side by side belonging to the genera Dacrydium, Podocarpus (Taxac.), Leptospermum (Myrtac.), Styphelia (Epacrid.), Coprosma (Rubiac.), Veronica (Scrophular.), and Celmisia (Compos.). In some cases the prostrate form is here hereditary, while in others it is unfixed and depends merely upon the station.

The combination of species forming the shrub steppe on the subalpine volcanic plateau, North Island, contains a considerable percentage of prostrate shrubs, some of which are more or less eréct under less xerophytic conditions.

Coastal rocks favour the prostrate form. Thus in such a situation near Island Bay, Wellington, there are Hymenanthera crassifolia Hook. f. (more or less hereditary), Coprosma Baueri Endl. (hereditary when juvenile), Veronica macroura Hook. f. var.* (perhaps hereditary when juvenile but erect when adult).

Other veronicas of coastal rocks are more or less prostrate, and this is strongly hereditary in V. chathamica Buchanan—so much so that a shoot grown vertically in a pot quickly assumed the horizontal direction (see Plate V, fig. 1).

An interesting instance of non-hereditary convergent epharmony of this growth-form is the wiry undergrowth of three species of Coprosma beneath the tussocks of Danthonia antarctica Hook. f. at some 250 m. altitude in Auckland Island. One of the species, C. foetidissima, is “normally” a tall twiggy shrub, and the other two are medium-sized divaricating shrubs.

6. Persistent Juvenile Forms.

About two hundred species of New Zealand vascular plants, belonging to thirty-seven families, show a more or less well-marked distinction between the juvenile and adult stages of development, while in perhaps one hundred species the differences are very great indeed. The most interesting cases are those in which a juvenile form remains permanent for a number of years, so that in its ontogeny the individual passes through two, or even more, distinct stages, and not infrequently through two

[Footnote] * I am inclined to think it would be better to consider this a species. It differs considerably from the typical form, which grows in the East Cape district.

[Footnote] † Heteroblastic development is a world-wide phenomenon which has not received nearly the attention it deserves from writers on evolution. It is its occurrence in so many endemic species in New Zealand that makes data from this region of special interest. In 1879 I. Bayley Balfour recorded a number of striking examples from the Island of Rodriquez—e.g., Clerodendron laciniatum Balf. f., reminding one of the New Zealand Nothopanax simplex Seem.; Pyrostria trilocularis Balf. f.; Fernelia buxifolia Lam., a rubiaceous plant, evidently when juvenile somewhat of the divaricating shrub form; and Mathurina penduliflora Balf. f. (Turnerac.), which has long narrow juvenile and broad adult leaves, as in Parsonsia heterophylla A. Cunn. and other New Zealand plants. Altogether seventeen species of trees and shrubs and one herb out of 175 species of spermophytes show marked dimorphism. Goebel (1889–93) gives a number of examples of heterophylly, &c., referring the phenomena in some instances to direct outer stimuli, and he deals further with the matter in his “Organography of Plants” (1900–5) and his “Experimentelle Morphologie” (1908). Diels (1906) goes into the matter at considerable length, using many important illustrations from his observations in Western Australia. As for the phenomenon in New Zealand, Hooker was the first to refer to it, in his splendid “Introductory Essay to the New Zealand Flora” (1853, p. 1). Kirk gives many details in his “Forest Flora” (1889), and these are supplemented by Cheeseman in his Manual. Finally, my own writings since 1899 contain a good deal of scattered information not previously published.

– 24 –

growth-forms. Although the juvenile and adult forms may be so distinct as to virtually represent different species, yet in many cases the adult does not appear suddenly, but intermediate stages occur. In these there is very frequently a combination of characters which are primarily quite distinct. Thus in the intermediate leaf-form of Parsonsia capsularis R. Br. (see fig. 2) there are all kinds of combinations between the early seedling short rounded leaf and the later long narrow one. Elaeocarpus Hookerianus Raoul also exhibits a remarkable series of leaf-combinations, for which see fig. 3. Further, there are transitions of general growth-forms, as when Sophora microphylla Ait. commences the adult stage with stout semi-erect but still flexuous stems. It seems clear from the above facts and from those that follow that the possibilities of both juvenile and adult are latent in the one plant, but each requires its necessary stimulus to set it free in its entirety. If the stimulus is not sufficient, then one or the other

Picture icon

Fig. 2.—Various Forms of Leaf in Parsonsia Heterophylla.
a, adult leaf; f and g, earliest form of leaf, but often more circular; d, e, and h, transitional forms; b and c, second type of juvenile leaf. Life size.

form may persist, or there may be a combination of characters, as in the transitional forms. In any case, heredity comes in, and this has attained to such a degree that under normal conditions there is a juvenile stage of a certain average duration, a transitional stage, and an adult. Different degrees of heredity have arisen, as I believe, in proportion to the length of time the original stimuli have functioned, combined with their intensity, and abnormal increase or decrease of stimulus can in many instances hasten or retard the procession of events. There is in some measure, perhaps, species-making going on before our eyes. This is best seen in those cases where the juvenile form produces flowers, for if progressive development should cease at this point what is virtually a species distinct from

– 25 –

the adult has appeared. Should such a flowering juvenile form be epharmonic, then, as Diels has shown, we are face to face with a case of ontogenetic evolution (1906). In some of the species the juvenile and adult forms can both clearly be shown to be epharmonic (e.g., Veronica lycopodioides Hook. f., Carmichaelia subulata T. Kirk, Discaria toumatou Raoul, Potamogeton Cheesemanii A. Bennett, Clematis afoliata Buchanan); they can even be experimentally produced or prolonged. In other cases epharmony can only be inferred (Sophora microphylla, Podocarpus dacrydioides, Rubus schmidelioides); and in others it is more or less obscure (Parsonsia heterophylla, Pseudopanax crassifolium C. Koch, Pittosporum patulum Hook. f.). There is, therefore, a gradual gradation from the known to the unknown, but, as the main features are alike throughout, it is reasonable to assume an epharmonic origin in most cases, notwithstanding that contradictory examples occur, and to consider that there is a relation between the age of the form and its relative stability. Here there is no attempt to go thoroughly into the phenomenon under consideration; certain typical examples are alone discussed.

The significance of the divaricating growth-form has been already noted. It may be remembered it is eminently xerophytic, extremely well defined, and present in various unrelated families. But this form is not confined to shrubs alone, but appears as a persistent juvenile stage in the life-history of certain plants, which are thus xerophytic shrubs for some years and finally ordinary mesophytic forest-trees. The following are examples: Pennantia corymbosa Forst. (Icacinac.), Hoheria angustifolia Raoul, Plagianthus betulinus A. Cunn. (Malvac.), Sophora microphylla Ait. (Legum.), Elaeocarpus Hookerianus Raoul (Elaeocarp.).

The case of Sophora microphylla Ait. is the most instructive. It must be considered along with the remaining species—S. tetraptera* J. Mill., S. grandiflora Salisb., and S. prostrata Buchanan. All the species commence with hypogeal cotyledons, and the first, or first two, leaves are simple and arrested structures, but the succeeding ones are pinnate and of the adult type. The primary stem is erect and somewhat flexuous (see Plate VI, fig. 2), except in the case of S. grandiflora. This species continues to grow erect, and in time develops into a small tree. There is no heterophylly beyond the early simple leaves, and no hint even of any xerophytic shrub stage. With S. microphylla the progress of events is very different. Here the early seedling soon develops into a xerophytic divaricating shrub, and so it will remain for some ten years or more, and attain a height of perhaps 1·4 m. before the more or less erect branches shoot upwards, the forerunners of the mesophytic tree form (see Plate III, fig. 1). It is quite common to see a specimen which is shrub at the base and tree above. Occasionally the upper part of the shrub form will blossom, but I do not think this ever happens before the tree itself flowers. Sophora prostrata never grows out of the shrub state; it is a fixed juvenile form, which, moreover, reproduces itself true from seed. Between S. microphylla, and S. prostrata there are intermediates. With regard to S. tetraptera, the juvenile plant differs but little from the adult (see-Plate VI, fig. 1), though it has for a time a few flexuous twigs. I have

[Footnote] * Under this name I include the Chatham Island plant, a form in the neighbourhood of Auckland City, and the Chilian plant. As for the Auckland plant, I do not know its juvenile state well enough to speak with certainty, but in any case its behaviour, if different from that stated, would not in any way affect my conclusions.

– 26 –

seen only one specimen raised from Chilian seed,* and it resembles closely the Chatham Island plant. S. microphylla and S. prostrata grow side by side at the lower Waimakariri Gorge, Canterbury Plain.

In the above case of Sophora the adult form is probably the stem form, and the xerophytic divaricating shrub form an epharmonic adaptation which arose during a probable period of drought on the east of the Southern Alps at the time of the glacial period (see Diels, 1896, and Cockayne, 1900). In certain parts of the problematical Greater New Zealand where the climate still remained sufficiently wet the ancestral Sophora would remain unchanged; so we still see S. grandiflora in the East Cape district and S. tetraptera in the Auckland district and the Chatham Islands. In the South Island there is only S. microphylla and S. prostrata, in the former of which the xerophytic stimulus never evoked an absolutely hereditary form, whereas in the latter the effect of the stimulus is much more deep-seated. To what extent such a stimulus can leave its mark is shown in the forest-tree Elaeocarpus Hookerianus, which at any age may put forth reversion shoots high up the trunk or on the branches. The heteromorphy in the other species listed above may be similarly explained. There is first of all a short-lived erect mesophytic stage, then a long-persisting xerophytic stage, and a final adult mesophytic stage. The first stage, suited as it is to shelter by ground-plants, &c., is epharmonic; it may also be considered a survival from the ancestral plant. The second (xerophytic) stage was epharmonic during the steppe-climate period of the eastern South Island, but is certainly beneficial no longer; and the adult stage is more or less a return to the original form, but now called forth by the present mesophytic conditions. According to this supposition, it is considered that the tendency to both xerophytic and mesophytic form is latent in the plant, and that one or the other will appear as soon as the necessary intensity of stimulus is reached. Until that is the case, whichever form is the more hereditary—i.e., the more strongly fixed—will persist, even though it is far from being epharmonic.

In a considerable number of instances there is a mesophytic juvenile stage and a xerophytic adult. In this class the present mesophytic conditions are not sufficient to inhibit the strongly hereditary xerophytic form, which also in a number of cases is in harmony with the xerophytic stations affected by these plants. The following examples of this and other persistent juvenile forms may be noted:—

(1.) Shrubs which are leafy as juveniles, but leafless as adults, when they have flat or terete green assimilating stems — e.g., species of Carmichaelia, Notospartium, and Corallospartium. How unstable really is the xerophily of many species of Carmichaelia is shown by their abundant production of leaves in shady stations.

(2.) Shrubs with an abundance of leaves, sometimes very thin, when juvenile, but of the cupressoid form when adult—e.g., certain Taxaceae (see Griffen, 1908), whipcord veronicas, and some species of Helichrysum belonging to the section Ozothamnus.

[Footnote] * The seed was very kindly sent to me by Dr. Eug. Autran, of Buenos Ayres, and the seedlings were raised by Mr. T. W. Adams, to whom I am greatly indebted.

[Footnote] † The divaricating form of Elaeocarpus Hookerianus and the juvenile Pseudopanax crassifolium, with its thick, narrow, stiff, deflexed leaves, certainly seem out of place in a rain forest, where they are assuredly not epharmonic structures.

[Footnote] ‡ Carmichaelia gracilis J. B. Armstrong is leafy in the adult; it is a scrambling iane, and grows in wet ground or swamps. C. grandiflora Hook. f. is deciduous, but abundantly leafy in spring and summer. C. odorata Colenso is also leafy.

Picture icon

Fig. 1.—Aristotelia Fruticosa
On left, juvenile; on right, adult
Fig. 2.—Pennantia Corymbosa.
On left, adult, on right, semi-juvenile in bloom

Picture icon

Pittosporum Divaricatum
Seedling of Westland form × 3

– 27 –

The juvenile stage in these plants, the Taxaceae excepted, does not usually persist for any long period, and may be compared to the first stage in Sophora and the various divaricating shrubs. But in the veronicas, as I have shown, it can be made to persist artificially for years, so long as the plant is kept in moist air. Further, reversion shoots are frequent even on such a typical xerophyte as Helichrysum Selago, while it seems probable that hereditary semi-juvenile races occur of Veronica tetragona‡ Hook. and V. lycopodioides* Hook. f. These are further dealt with on p. 45. In the case of Dacrydium laxifolium Hook. f., a prostrate or suberect mountain-shrub, growing in wet ground or bogs, it is quite common to find juvenile plants with the lax spreading leaves blooming freely, and never developing into the cupressoid adult. The juvenile stage of D. intermedium T. Kirk frequently becomes a tree, and flowers and fruits as abundantly as the “normal” adult growing in the same swamp forest. This flowering juvenile was given the varietal name gracilis by Kirk.

(3.) Nearly all the divaricating shrubs have a primary juvenile mesophytic stage. This is generally but transitory, but I have already shown in the case of Pittosporum divaricatum and Corokia Cotoneaster how the early stage may persist in the forest and reach its full stature. Semijuvenile plants of the Pittosporum may also flower.

Aristotelia fruticosa Hook. f. (Elaeocarp.) is an interesting case. The early seedling is erect, mesophytic, and, compared with the adult, shows a most remarkable variety of leaves. These are often more or less lanceolate, toothed, lobed, or pinnatifid (see Plate VII, fig. 1). Later on the divaricating form appears, which may finally be of the most intense character, the small frequently more or less oblong leaves being scanty, and the ultimate shoots almost spinous. But this form is not truly stable, plants growing in an adjacent Nothofagus forest being much more mesophytic. Even when quite in the open, there are forms still divaricating to some extent, it is true, but juvenile so far as leaf-form goes, and these develop no further, and blossom. This semi-juvenile fixed form should be considered older than the “normal” adult, and it may represent the pre-glacial plant.

Suttonia divaricata Hook. f., so far as I have observed, has no seedling mesophytic stage. But even this “well-fixed” species when growing on the Poor Knights Islands has leaves three times the size of those of the usual stations.

(4.) Nothopanax simplex Seem. and N. Edgerleyi Harms. (Araliac.) have also a mesophytic juvenile form, but the adult must be considered mesophytic likewise. Both are rain-forest plants, while the former is found also in certain subalpine scrub. I have not full details regarding N. Edgerleyi, the juvenile form of which sometimes so closely resembles that of N. simplex that I, for one, cannot distinguish between them, so my remarks are confined to the latter species. The early stage has a fern-like, much-cut, thin and large leaf. This is succeeded by a second stage with ternate

[Footnote] * Cheeseman found a semi-juvenile form of V. tetragona at the base of Tongariro and Ruapehu, and writes (1908, p. 281), “Probably it is an intermediate state between the juvenile stage and the fully matured one, but if so it must persist for many years.” Mr. Poppelwell collected a form of whipcord Veronica on the Garvie Mountains, a plant of which has kept the semi-juvenile form for two years in my garden. So, too, from some notes sent to me by Mr. F. G. Gibbs it is evident that he has had in cultivation a very similar plant.

[Footnote] † Hymenanthera dentata R. Br. var. alpina T. Kirk also develops semi-spines under very dry conditions.

– 28 –

leaves, and this by the simple- and thicker-leaved adult. In some localities the much-cut form is suppressed to some extent, or almost entirely absent (Auckland Islands; but see Cockayne, 1904, p. 249,* and pl. 11). The closely related Nothopanax parvum Cockayne also seems to lack a cut-leaved stage. N. anomalum Seem., although frequently a forest-plant, has a juvenile mesophytic form with small ternate leaves and an adult divaricating shrub form connecting the ternate-leaved form of the genus with the divaricating shrubs.

(5.) In this class come a considerable number of plants which cannot with any confidence be referred in their different stages to special outer factors. Take the case of certain species of Pseudopanax (Araliac.): two (P. crassifolium C. Koch and P. ferox T. Kirk) have the curious narrow deflexed juvenile leaves and unbranched stem, but in P. lineare C. Koch, a subalpine shrub, the virtually similar juvenile leaves are erect; and in P. chathamicum T. Kirk they are wanting altogether, the juvenile and adult leaves not being very different.

The primary seedling leaves of P. crassifolium are somewhat similar in form to the adult, but, of course, much smaller. They are erect, and never deflexed. P. ferox, on the contrary, commences with narrow-linear toothed leaves of the second stage, which are not erect, but horizontal for a time.

The small-leaved juvenile and the large-leaved adult forms of the root-climbing fern Blechnum filiforme Ettingsh. cannot be explained epharmonically, though there probably is, or has been, some relation of the sort, since the first-named is the common ground form (creeping form) and the large-leaved the climbing form. Nor can I suggest any explanation of the two juvenile leaf-forms of Parsonsia heterophylla and P. capsularis R. Br. (see fig. 2). In the former species the long narrow-leaved shoots occasionally flower, and in the latter there is a fixed flowering juvenile race occurring in the uplands of the South Island which I consider a distinct species.

Weinmannia racemosa L. f. and W. sylvicola Sol. (Cunoniac.) are two closely related species whose flowers are virtually identical, and which differ merely in the adult leaf of the first-named being entire and of the other compound. The early seedlings of both are identical; they are erect, their leaves are simple, toothed, thin, and hairy. Then comes a second stage, in which in W. racemosa the leaves are ternate, and in W. sylvicola both ternate or pinnate. At this stage, when both plants are merely bushy shrubs, they can flower, and need not develop into trees. Frequently on the heath lands of northern Auckland W. sylvicola attains 3–4 m. in height; the leaves are large, and have many leaflets, yellowish in colour, and although Mr. H. Carse, myself, and others have seen hundreds of these tall juvenile plants we have never seen them in flower. Ackama rosaefolia A. Cunn. (Cunon.), if not actually a companion plant, grows near by on the forest's outskirts, &c., and its adult form so much resembles this juvenile Weinmannia that no one could distinguish flowerless examples one from the other without a knowledge of certain quite obscure differences. The adults of the two species of Weinmannia are lofty forest-trees. From the above it seems reasonable to conclude that W. sylvicola is merely a fixed

[Footnote] * Through a clerical error “entire-leaved” is printed several times instead of “simple leaves.” The leaves are more or less serrate, but compared with the juvenile they are virtually “entire.”

[Footnote] † The distinctions given by Kirk in the “Forest Flora,” p. 113, do not hold in practice, so far as the leaf is concerned.

– 29 –

juvenile stage of W. racemosa, or else that the former is the stem form and W. racemosa a mutation or an epharmonic variant that has become fixed.

Several instances of juvenile blossoming have already been given. The following are additional examples:—Ranunculus Lyallii Hook. f. (the juvenile has a reniform leaf and the adult a peltate; reversion leaves occur as a result of bad nutrition; there are intermediates between the two types of leaves): Pittosporum tenuifolium Banks & Sol. (the juvenile seems to me identical with P. nigrescens Hort.,* the plant so much used in certain parts of New Zealand for—hedges; as a hedge-plant, the juvenile form is alone to be seen, it being preserved by the constant cutting): Glematis indivisa Willd.: Dracophyllum arboreum Cockayne: Agathis australis Salisb.: Nothopanax Edgerleyi Harms. (one semi-juvenile form blooms and is the var. serratum T. Kirk): and Anisotome filifolia Cockayne and Laing. There are also a number of forest-trees which remain in the shrub stage and flower (see Cockayne, 1908, p. 22).

Each of the above cases' would need deciding on its merits as to whether the flowering juvenile might be the beginning of a new line of descent, or was merely a reversion. I will only discuss the case of Anisotome filifolia Cockayne and Laing.

This is an herb with the leaves in an erect rosette and a long tap-root which grows upon stony debris where there is a steppe climate in the mountains of Nelson, Marlborough, and Canterbury. The leaves are grassy, some 20 cm. long, ternately divided into segments which are filiform if the plant grows in the open, but 3 mm. broad, or broader, when growing in the shade. Both forms produce flowers. Seedlings raised from the filiform xerophytic form had broad segments (see fig. 38, pl. 12, in Cockayne, 1900, and also pp. 295–97). The broad leaves are certainly beneficial for promoting rapid growth in a dry station, nor will the seedling be exposed to as rigorous surroundings as the adult, protected as it will be by the stones. Its form is therefore epharmonic. The broad-leaved adult of the shade is then a flowering juvenile, which may or may not be “fixed,” but, if fixed, it would be an example of ontogenetic evolution, the arrival of the new species dating from the first time the juvenile plant reproduced its like from seed.

Many of these heteroblastic species put forth when adult typical juvenile or semi-juvenile shoots, as the case may be. Such may often be traced to a special stimulus. Thus, stems of Phyllocladus alpinus Hook. f. when prostrate on wet soil may bear abundance of true leaves, but those in a drier position have phylloclades only; Discaria toumatou Raoul cropped by rabbits produces leafy shoots only; and Ranunculus Lyallii Hook, f. grown in dry soil under unfavourable conditions may develop a certain number of reniform seedling leaves.

The position of the reversion shoots upon the plant differs in different species. Very often they are confined to near the base, in which case they may be merely developed resting buds. Pseudopanax crassifolium C. Koch and Weinmannia racemosa L. f. when cut to the ground regenerate from the stump by means of juvenile shoots. Pittosporum tenuifolium Banks & Sol., as a hedge-plant, remains permanently juvenile through frequent

[Footnote] * H. M. Hall (1910) is of the same opinion.

[Footnote] † Other species of Pittosporum also occur at times in these hedges through the sowing of mixed seed, and so other forms of leaf may be occasionally present.

[Footnote] ‡ I noted one adult plant growing on a sand-dune that was almost, if not entirely, without spines, the xerophytic station notwithstanding.

– 30 –

clipping. The shrubby bases so frequently seen of Pennantia corymbosa Forst., Hoheria angustifolia Raoul, &c. above which the flowering and quite different adult rises, are not very long-lived, but finally die and are cast off. In some cases the distinction between juvenile and adult is equally great, as in the above, but the stability of each form is weaker, and the power of the cell derived through heredity to produce one or the other is present in every shoot, no matter how far from the base, reminding one somewhat of the behaviour of a “graft hybrid.” Examples are: Dracophyllum arboreum Cockayne, whipcord veronicas, Podocarpus dacrydioides A. Rich, Aristotelia fruticosa Hook. f. In Elaeocarpus Hookerianus Raoul reversion shoots occur high up the tree, but I have not noted them in the uppermost branches. In these last-cited examples an observable stimulus does not seem necessary to bring forth the special form; it is rather as if very little indeed — probably some slight internal change — can suffice to upset the equilibrium of the cell upon' which one or the other form depends. An analogous example is a variegated form of Veronica salicifolia which originated spontaneously in the garden of the late Mr. W. Gray, of Governor's Bay, for many years an enthusiastic cultivator of New Zealand plants. The first leaves of each shoot have an irregular band of green down the centre of every leaf, but as these become older chlorophyll gradually invades the pale portion until the leaf becomes normally green. Shade leaves are at first without any chlorophyll.

Picture icon

Fig. 3. — Leaf-forms of Elaeocarpus
a, small adult leaf; b, transition to adult; c and d, early long narrow form: e, f, and g, early obovate short form. The long narrow and short obovate or rotund leaves are associated with divaricating branching. Life size.

VI. Hybridization.

Hardly anything is known as to the occurrence of wild hybrids in New Zealand. But field observations on this head are, in any case, merely suggestive, and, at most, pave the way for experiment.

Long ago hybrids were raised in cultivation by Mr. Anderson Henry and others in Great Britain from some of the large-leaved lowland species of Veronica. What I take to be hybrids—one especially from V. pimeleoides Hook. f.—have originated spontaneously in the semi-wild collection of indigenous plants in the Christchurch Domain. Mr. D. L. Poppelwell has sent me a hybrid from his garden which he considers V. salicifolia × V. decumbens. It is somewhat of the salicifolia type, but with small glossy leaves; I have not seen the flowers. Recently Mr. A. Lindsay, of Edinburgh has raised one or two hybrids of which the parents are known. The

– 31 –

most important of these is V. Hectori Hook. f. (a “whipcord veronica”) × V. pimeleoides Hook. f. (a small glaucous-leaved straggling rock-plant with blue flowers), and the result is a plant said to be identical with or near to V. epacridea Hook. f. If this is true, it opens up much suspicion as to the validity of many species of the genus in New Zealand, and, at any rate, in the case of variation in general, as some of the species are gynodioecious,* hybridism may be the simple explanation.

Mr. McIntyre, who had charge of the famous collection of New Zealand plants of the late Mr. H. J. Matthews, raised a good many hybrid forms of Celmisia, all of which appeared to have the so-called C. verbascifolia as one of the parents. I have seen a Celmisia on Jack's Pass which was most likely a hybrid between C. spectabilis and C. coriacea. Also, C. mollis Cockayne is possibly of hybrid origin, with C. spectabilis as a parent. In short, hybridization may account for some of the variation in Celmisia. Acaena, again, is a very variable genus, which suggests hybridization. Buchanan was the first to call attention to this matter, and he described a supposed hybrid between A. Sanguisorbae Vahl. and the introduced A. ovina A. Cunn. (1871, p. 208). Kirk reduced this to var. ambigua of A. ovina, notwithstanding that the inflorescence is altogether different from that of that species. Bitter (1911, pp. 297–321) describes fifteen hybrid forms of Acaena, illustrated by figures of leaves, in which varieties of A. Sanguisorbae, A. microphylla, and A. glabra are parents, one or the other. ‘These forms have originated spontaneously in the Bremen and other Continental botanical gardens. Bitter is convinced they are true hybrids, and that the only question that can be raised is as to the parentage that he suggests for them. A full account is given of each form.

I have seen, judging from the capsule, what appear to be wild hybrids between Phormium tenax Forst. and P. Cookianum Le Jolis. A good deal of the variation in P. tenax may be due to hybrid elementary species, for that it is made up of many such entities seems very probable.

Melicope Mantellii Buchanan is supposed by some to be a hybrid between M. simplex A. Cunn. and M. ternata Forst. (see Kirk, 1889, p. 118). I have proved that it comes true from seed, and in the absence of experimental evidence it is quite as reasonable to suppose it is an elementary species connecting M. simplex and M. ternata. All three have ternate juvenile leaves; M. ternata remains at this stage but with much larger adult leaves, M. Mantellii has both simple and ternate leaves in the adult, and M. simplex is a divaricating shrub when adult with simple leaves.

VII. The Struggle for Existence.

Plant-ecologists have many opportunities for observing. Various phases of the struggle for existence. They have also some opportunity of judging

[Footnote] * I am indebted to Professor I. Bayley Balfour, F.R.S., for calling my attention to this phenomenon in our veronicas, which he was the first to discover. I had previously wondered why certain species in my garden never produced seed, and others very little, and had ascribed it to the absence of the proper pollinating insect. How far the phenomenon is present in wild plants has not been as yet ascertained.

[Footnote] † Probably C. verbascifolia Hook. f. = C. Brownii Chapman.

[Footnote] ‡ The Chatham Island form, with its thin broad leaves, is distinct, so far as I know, from any of the mainland forms.

– 32 –

as to the likelihood of extremely small* variations being preserved or the contrary. It must be understood that the “struggle “is not only between the individuals of the competing species, but also between these and their environment. This was distinctly stated by Darwin, who refers to the struggle for life against the drought on the edge of a desert (1899, p. 46). In many instances this struggle with outer circumstances is the more important; it is also the deciding factor as to what plant-form can gain a footing in the first instance.

The formations themselves offer various conditions according as they are “open” or “closed,” for in the former there is apparently room for new-comers, whereas in the latter it is almost impossible for a species from without to gain admittance. This fact is of major importance, for, amongst other matters, it has a strong bearing on the much-debated question regarding former land connections with distant islands as opposed to bird carriage, &c., across wide areas of ocean. The case of New Zealand as a whole is of great interest in this regard, especially as many misstatements have crept into evolutionary writings regarding the spread of the introduced plants and their rapid “replacement” of the indigenous flora. I will state briefly what I believe to be the true state of affairs.

There have been recorded for New Zealand up to the present some 555 species of introduced plants, but less than 180 can be considered common, whilst others are local, rare, or even not truly established as wild plants. Many at first sight appear better suited to the soil and climate than are the indigenous species, and over much of the land they give the characteristic stamp to the vegetation; but this is only the case where draining, cultivation, constant burning of forest, scrub, and tussock, and the grazing of a multitude of domestic animals have made absolutely new edaphic conditions which approximate to those of Europe, and where it is no wonder that the European

[Footnote] * During the discussion following the reading of this paper the expression “extremely small “was criticized as not giving a fair representation of the views of Darwin and his followers. Darwin, however, writes (1889)—p. 45, “Variations, however slight “; p. 58, “any advantage, however slight “; p. 59, “extremely slight modifications “; p. 69, “individual differences, too slight to be appreciated by us.” Weismann puts the case more strongly still (1910, p. 25): “For the question is not merely whether finished adaptations have selection value, but whether the first beginnings of these, and whether the small, I might almost say minimal, increments which have led up from these beginnings to the perfect adaptation have also had selection value.” Wallace, on the other hand (1889, pp. 126,127), claims that though Darwin used the word “slight “and “small amount,” these terms are “hardly justified,” since the variability of many important species is of considerable amount, and may very often be properly described as large.

[Footnote] † The case of Krakatoa, important as it otherwise is, seems to me to have but little bearing on this question, since the distance from the mainland is too trifling.

[Footnote] ‡ Wallace (1889, pp. 28, 29) refers to Trifolium repens exterminating Phormium tenax; excellent pasture destroyed in three years by Hypochoeris radicata, which can even drive out white clover; and Sonchus oleraceus growing all over the country up to an elevation of 6,000ft. Kirk (1896. p. 18) not only attributes the “displacement” of Phormium to grasses and clovers, but also Mariscus ustulatus, and even Pteridium esculentum (bracken fern). Further on (p. 19) he states that Aciphylla Colensoi is gradually replaced by self-sown pasturage plants. However, he also calls attention to the effect of grazing and trampling by cattle and horses as aiding the plants in their work, which, of course, is a very different matter from the effect of plants alone.

– 33 –

invader can replace the aboriginal.* On the other hand, although this foreign host is present in its millions, and notwithstanding abundant winds and land-birds, the indigenous vegetation is still virgin and the introduced plants altogether absent where grazing animals have no access and where fires have never been. On certain subalpine herb-fields the indigenous form of the dandelion (Taraxacum officinale Wigg.) is abundant, and yet the introduced form, with its readily wind — borne fruit, has not gained a foothold, nor even the abundant Hypochoeris radicata L., though it may be in thousands on the neighbouring tussock pasture, less than one mile away. On Auckland Island introduced plants occur only in the neighbourhood of the depots for castaways, but on Enderby Island, where there are cattle, they are much more widely spread. Even where the rain forest has been felled or burnt, and cattle, &c., are kept away, it is gradually replaced by indigenous trees and shrubs—i.e., in localities where the rainfall is sufficient.

Some of the indigenous species are quite as aggressive, or even more so, than any of the introduced. In primeval New Zealand each would have its place in the association to which it belonged—there would be no aggression; but when the balance of nature was upset by the fire or cultivation of Maori or European, then the plants best equipped for occupying the new ground become dominant, their “adaptations” for that purpose fortuitously present. The miles on miles of Leptospermum scoparium and Pteridium esculentum were absent in primitive New Zealand. So, too, the pastures of Danthonia semiannularis R. Br. in Marlborough, and the many acres of Chrysobactron Hookeri Colenso (Liliac.) in the lower mountain region of Canterbury. Celmisia spectabilis Hook. f., an apparently highly specialized herb for alpine fell-field or tussock-steppe conditions, is now on the increase in many montane parts of the Ashburton-Rakaia mountains and valleys, owing to its being able to withstand fire, the buds being protected by a close investment of wet decayed leaf-sheaths.

Nor are all the introduced species aggressive, by any means. Some can barely hold their own; others are limited to certain edaphic conditions. Thus, Glaucium flavum Crantz occurs, as yet, only on the coast of Wellington, chiefly in the neighbourhood of Cook Strait. It is confined to gravelly or stony shores, and appears unable to grow on the clay hillside. And yet where the latter is, in one place near Lyall Bay, covered with gravel there is a large colony of the plant, whence none have found their way on to the adjacent hillside. Lupinus arboreus, now so common on New Zealand dunes, appears unable to spread beyond the sandy ground.

The often-quoted stories (see footnote, p. 32) of white clover (Trifolium repens L.) being able to wipe out Phormium tenax, of Salix babylonica overcoming the watercress (Radicula Nasturtium-aquaticum), of Hypochoeris

[Footnote] * New Zealand may be roughly divided into three areas—viz., the cultivated, the pasture lands, and the primitive. It is only in the pasture lands that a real struggle between the introduced and the indigenous plants is taking place, and even there the contest is very unequal, through the grazing, burning, and seed-sowing factors. Many pastures, however, are altogether new associations, as in the case of forest being felled, then burned, and the ground sown with grasses, &c., even before the ashes of the trees are cooled, so that at once there is a foreign pasture brought into existence and subject to an entirely new set of conditions from that which governed the forest. This is certainly not biological “replacement.”

[Footnote] † Introduced, not native, birds.

[Footnote] ‡ The species may be D. pilosa, but I have no specimens for identification 2—Trans.

– 34 –

radicata displacing every other plant of excellent pastures in Nelson, are without foundation. P. tenax has certainly been eradicated in many places, and perhaps, in a sense, replaced by white clover, but not until fire and feeding of stock had killed the plant.

The great screes, called locally “shingle-slips,” which are such a characteristic feature of mountain scenery in much of the South Island, possess a most scanty and scattered vegetation, made up of some twenty-five highly specialized species belonging to thirteen families, of which twenty species occur in no other formation. Here the struggle between the individuals is nil, but that with the environment, especially the unstable substratum, is most severe. I know of no instance where a non-indigenous plant has established itself on a true alpine shingle-slip.* In such a station no plant could gain a footing unless provided beforehand with some special “adaptations “fitting it for the severe conditions. The shingleslip association, moreover, is neither the climax of a succession nor is it part of such; it is an association complete in itself, and connected with no other. Of a number of plants germinating by chance on a shingle-slip, the seedling which possessed a slightly more xerophytic structure than its fellows would be none the better, but would perish equally. Granting that natural selection can intensify characters by slow degrees, the conditions would select too rigorously—there would be no survivors. It is almost equally difficult to see how epharmony could work, either. A plant to gain a shingleslip must come from some specially xerophytic station. This is shown by the presence of Veronica epacridea Hook. f. and V. tetrasticha Hook, f., rock - xerophytes. Perhaps the true shingle - slip plant Craspedia alpina Backh., a summer-green herb with leaves in rosettes and thickly covered with, a deep snow-white wool, also arrived from some other formation, and its abundant wool and deciduous leaves have arisen epharmonically. The dimorphic succulent Claytonia australasica Hook. f. also occurs elsewhere, one form being found in cold streams and damp gravel. Its rapid response to a xerophytic stimulus accounts for its presence.

The seedlings of the true shingle-slip plants are, so far.as they have been studied, strongly xerophytic at an early age. Thus an examination of a seedling of Stellaria Roughii Hook. f. raised by me under mesophytic conditions showed, “in the' elastic stem, pale glaucous-green leaves, and early succulence of the seedling, how hereditary are the most striking peculiarities of shingle-slip plants “(Cockayne, 1901, pp. 267–69).

An interesting point is the occurrence of two distinct species of Cotula, or varieties of one species, it matters not, which are epharmonically equivalent. Taxonomically they differ in colour of florets, size of flower-head, and size of involucre as compared with head. Accumulative selection could do nothing here; both plants thrive equally well, and there is no competition except with the environment. Mutation alone can explain this remarkable case, or some cause unknown. Another some what similar example is Notothlaspi rosulatum Hook, f. and N. australe Hook. f. and its var. stellatum T. Kirk. Anisotome carnosula is in appearance exactly like A. diversifolia Cockayne, but there are technical differences

[Footnote] * Introduced plants occur at times on small screes at base of rocks, and on riverterrace scree in the lower mountain belt.

[Footnote] † Weismann writes (1910, p. 61), “How often has the senseless objection being urged against selection that it can create nothing; it can only reject…. But in rejecting one thing it preserves another, intensifies it, combines it, and in this way creates what was new.

– 35 –

in the umbel and the involucral bracts. A. diversifolia has been found so far on only one mountain on which A. carnosula is not known to occur; but the species are so much alike that they could only be recognized when in bloom and examined closely.

To trace the evolution of the shingle-slip plants it seems clear that one must go back to the origin of the shingle-slips themselves from their small beginnings before the eastern peaks of the Southern Alps were disintegrated into rounded summits. If for any reason the climate were wetter,* there would be a similar condition of affairs to what governs the shingle-slips of Westland to-day where true shingle-slip species are absent. On the embryonic debris slopes many plants could settle down, and to the believer in natural selection nothing could appear more probable than for these to have been gradually changed in accordance with the slowly changing environment, species after species going to the wall, until only the few highly differentiated should remain. Even these are absent over wide areas of the most extensive and unstable of these alpine deserts.

An exactly similar argument to the above would apply to water formations, especially as there are cases where true water-plants — e.g., Potamogeton Cheesemanii Bennett—flourish in situations where they are quite uncovered for considerable periods. Even for unstable dunes, where there is certainly no struggle between plant and plant, and where no non-sand-binding form could possibly become established, a similar argument would apply, since all degrees of sand-movement exist in a dune-area. But in all the above cases we do know that ecological factors can evoke structures such as are essential, and we do not know for a fact that selection can intensify a character beyond a certain limit. In the tussock - grass Poa caespitosa the power to respond to sand-movement is already present, although its adaptations fit it for other conditions; thus it has occupied the recent drifting sands of Central Otago. Cases such as these, of stony debris, water, and dune, should be decided not on preconceived opinions or theories, but on the most reasonable conclusions from the observed facts.

Rock-vegetation, although open, affords plenty of scope for the struggle for existence both between the individuals and with the environment, since, leaving the lithophytes out of the question, the space for rock-crevice plants is very limited.

On the recent roches moutonnées alongside the Franz Josef Glacier the occupation of rock is now in progress. The pioneer plant is a dark-coloured species of moss, which when it happens to grow in a crevice forms a soil, an essential for the successful germination of seeds in such a station. The first-comers are all plants of some neighbouring association, mostly xerophytes, some herbs, and other shrubs, or even trees, whose long roots can penetrate into the chinks. Exceptions to this are the filmy fern Hymenophyllum multifidum Swz., the epiphytic or rock-dwelling orchid Earina autumnalis Hook. f., and Lycopodium varium R. Br.; but it must be remembered the atmosphere is nearly always saturated with water-vapour. The above first-comers react one upon another, the most vigorous finally conquering; but this vigour depends rather upon age than on greater

[Footnote] * Speight, in a carefully considered paper (1911), brings forward a good deal of very suggestive evidence as to the probability of a wetter climate on the east than the present one following the steppe climate. The most important fact adduced is the former presence of extensive forests where steppe alone now exists, since such forest could only be established during a period with many rainy days, and no other explanation seems to fit the case.

– 36 –

suitability for the station. At any rate, the chance for natural selection to effect anything here is very remote, although the competition is powerful.

The number of true rock-plants in New Zealand is comparatively small; but, on the other hand, a great many xerophytes, and even mesophytes, are encountered on dry rocks, but the latter are epharmonically modified during their individual development.

Even hygrophytes may gain a footing, as already seen in the case of Hymenophyllum multifidum. The most striking and truly amazing case is that of the kidney-fern (Trichomanes reniforme Forst. f.) and Hymenophyllum sanguinolentum Sw., which grow in the full blaze of the sun upon the lava of Rangitoto Island, Auckland Harbour. The fronds of both, as I saw them on a hot summer's day, were dry and curled up so as to appear dead, but Mr. Cheeseman informs me that in winter the kidneyfern covers the rocks with its translucent fronds, and that those of summer are not dead at all. It seems evident that in this case the protoplasm of these ferns must behave similarly to that of many lichens, and this will be an epharmonic adaptation. The question arises, does such a power lie latent in these ferns as normal rain-forest plants, ferns which cannot tolerate a drying wind or a hot sun and dry atmosphere; and, if so, how can it have possibly come about ? Probably the porous rock contains a good deal of water, and the air is usually not dry. Although I do not think that any modification through the struggle for existence takes place amongst rock-plants, yet this case shows that one cannot tell but that the most unlikely species might settle in certain stations, and so inaugurate a new line of descent, no matter how the evolution be brought about.

In closed formations the struggle for existence between individuals is very keen. As I write, in my garden, in a bed crowded with indigenous plants, two rapidly growing and far-spreading Chatham Island herbs have encountered, and one (Pratia arenaria Hook. f.) is rapidly replacing the other (Cotula Muelleri T. Kirk), a happening quite in accordance with the fact that the former plant is one of the most widely spread of the Chatham Island plants. Reduced to its ultimate factors, the struggle is chiefly one for nutriment in its widest sense, as Clements has shown (1905, p. 286); there is little actual destruction of one plant by another, though they function indirectly by cutting off light, using up nutritive salts, &c. In some cases the greater part of the struggle takes place amongst the young plants, and it is on their adaptations, which may differ much from those of the adult, that the establishment of the latter depends. This is specially evident in those heteroblastic species already dealt with which have ecologically different forms in their different stages. In a forest the conditions for the seedling and sapling trees are very different from those to which the adults are exposed. A favourable variation which might preserve a seedling in the struggle with its environment would possibly have little to do with the imperative demands of the adult. Small outward modifications of a very few individuals could hardly be preserved in the dense growth of saplings* in an upland forest of Nothofagus cliffortioides Oerst. The chief requisite of success here is rapidity of growth, a physiological

[Footnote] * The saplings may grow so closely that one can hardly force a passage through them.

[Footnote] † The case described in my little book, “New Zealand Plants and their Story,” of a species of Eucalyptus overcoming the eminently aggressive Leptospermum scoparium, through its more rapid growth, both germinating at the same time, is instructive in this regard

– 37 –

characteristic that, however much intensified, could bring about no specific differences unless correlated with structural change. In point of fact, the deciding factor in the struggle amongst a close-growing mass of these tree seedlings is probably age. Could all commence on exactly the same footing, then the determining factor would be the situation with regard to the food-supply and the illumination, and no slight beneficial modification would count in comparison.

As for the adult forest-trees, each has, as a rule, its own rooting-place, and its death depends chiefly upon its age, partly upon some disease or other, and but little upon the superior adaptations of its neighbour. Its growth-form, certainly, does have something to do with its longevity, as where spreading branches favour the presence of abundant epiphytes, whose weight may lead to damage and permit the attack of fungi.

A mixed rain forest, apart from modifications due to the nature of the topography, might be expected to offer constant conditions extending over a considerable period. But this is not so; topographically similar parts of a forest may show dissimilar undergrowth, the result of conditions which, similar at first, become dissimilar as the vegetation develops. Thus in the Waipoua Kauri Forest, of which I made a special study, a state of change ruled. In one part there was little undergrowth, and in another part such in abundance. This latter, in time, will, through survival of the fittest, change into forest with little undergrowth. These are two climaxes, and are expressions of the light factor, the dense undergrowth denoting the maximum and the final open forest with the close roof the minimum of illumination. Between these two climaxes there are many transitions. Bring in more light still and so increase the xerophily, the hygrophytes will go to the wall, until, with excess of light, a transition forest and finally a Leptospermum heath will be established (Cockayne, 1908, p. 30). From the above it follows that, even were natural selection at work amongst the young plants of any species, owing to the varying change of conditions brought about by these plants themselves there would be an insufficient length of time for any more suitable variety to arise, or, if such selection were very rapid, different types would be selected within a quite limited area. The believer in the efficacy of epharmonic variation would say that forest-trees have arisen from shrubs, or vice versa, owing to the stimulus of edaphic, climatic, and other factors, and that selection operated by eliminating those individuals which did not respond epharmonically at various stages of the plants' development. And the special evidence put forth would be that many species possess an unfixed epharmonic tree form and shrub form, while it is known that stature and other features can be modified through changes in nutrition. This, after all, is only Darwinian selection plus an assigned cause for rapid and sometimes favourable modification*; but it is far from being neo-Darwinian selection.

VIII. Distribution Of Species.

1. Distribution in General.

The distribution of species is primarily a matter of epharmony. Such, however, must in certain cases be referred to a state of affairs no longer

[Footnote] * I do not mean to infer that all modification is favourable.

– 38 –

present,* as in various instances of restricted distribution. Heat is a factor of prime importance, and, so, many species have a definite southern or altitudinal limit beyond which they do not extend (e.g., Agathis, Ipomaea, Veronica elliptica, Knightia, Senecio rotundifolius, &c.). This is not because they cannot exist quite well farther to the south, or at a higher altitude, but that on approaching their heat-minimum they cannot compete with the other better epharmonically suited competitors. Further, changes of land-surface have affected distribution in some cases, especially where they have caused permanent or temporary barriers.

The annual number of rainy days is also a most important controlling factor, and one whose effect is more plainly to be seen than that of heat. The densely forested west of the South Island and the sparsely wooded country beyond the average limit of the western rainfall to the east of the main divide stand out in startling contrast. On the west the evergreen canopy tree, and on the east the brown grass tussock, reflect in their respective dominance the prevailing ecological conditions. The slight differences, too, of the closely related Gaya Lyallii Hook. f. and G. ribifolia Cockayne are excellent examples of quite small but distinctly epharmonic distinctions influencing distribution.

Wind is another most powerful factor in New Zealand. According to their relative wind-tolerating power do certain shrubs, &c., replace one another on the shores of Paterson's Inlet, Stewart Island, so that the shoreline has become in its vegetation an exact index of the frequency and velocity of the wind. The above steppe district in the centre and east of the South Island is governed quite as much by the wind as by its moderate rainfall.

Quite common plants are extremely rare in certain localities. Cordyline australis Hook. f., a tree of physiognomic importance in many parts of both the North and South Islands, occurs in only one locality in Stewart Island. Leptospermum scoparium, usually so abundant, is represented by but one or two individuals in the Chatham Islands, where there is the ideal station for it to form a heath. The tree-fern Hemitelia Smithii Hook., so abundant in Stewart Island, is confined, so far as known, to one gully in Auckland Island. Psychrophyton eximium Beauverd is abundant on low alpine rocks on Mount Torlesse, Canterbury, but is wanting in similar stations on the range on the opposite side of the valley.

In some cases there is evidence that a plant has been much more abundant, but has been replaced by another species. This is true “replacement,” and very different from the so-called replacement of indigenous by introduced plants. Podocarpus spicatus R. Br. was an important member of the Stewart Island forest, say, five hundred to a thousand years ago. At the present time there remain only a few trees of that species, but it is common to find old trunks of this taxad on which are growing

[Footnote] * It is plain from the very nature of the case that perfect harmony can never be established between the growth-forms and the habitat, since change, progressive or retrogressive, is a feature of all formations, and growth-forms once epharmonic will persist long after their epharmonic relation is weakened or destroyed.

[Footnote] † G. Lyallii has larger, thinner, and much less hairy leaves than the eastern G. ribifolia. They have also drip-tips, which are frequently strongly developed. The juvenile forms are somewhat similar in the two trees.

[Footnote] ‡ This term “heath “I have used in my writings in default of a better, well knowing the formation is not truly analogous, except after fire, with the heaths of Europe. By the settlers, when full grown, it is known as “manuka” or “tea-tree scrub.” At this stage it is rather forest than heath.

– 39 –

full-sized trees of Weinmannia racemosa L. f. Sophora tetraptera J. Mill. is restricted in Chatham Island to the forest on limestone near the shore of the Te Whanga Lagoon, though elsewhere in New Zealand it can grow abundantly on rock similar to that of the rest of Chatham Island. The accompanying trees are the same in the limestone forest as in forest of the island generally, but it is evident the volcanic rock of the remainder of the island favours the other trees, which do not allow Sophora to become established. Or it may be that Sophora is a comparatively recent arrival.*

The distribution of certain species shows that epharmony is by no means so complete between plant and habitat in some cases as one might expect; or, in other words, that a plant can live in a position for which it is not perfectly fitted. Thus, Mr. R. G. Robinson, Superintending Nurseryman for the South Island, informs me that the dominant tree of the Tapanui Forest, Nothofagus Menziesii Oerst., cannot be grown in the adjacent State nursery, although N. fusca Oerst., a comparatively rare plant in that locality, can be grown with extreme ease; and yet I have seen N. Menziesii growing quite well on the flanks of Ruapehu as an isolated tree in the open. The slow growth of many indigenous trees as compared with introduced species is another case in point. On Antipodes Island the plant-associations are not distinguished by their different floristic components so much as by the relative abundance of the different species. This word “abundance “shows that all are not equally suited for each station, but that if a plant settles down on ground not specially fitted for its requirements it may be able to hold its place, the struggle for existence notwithstanding. So, too, with various stations on the Auckland Islands. A highly specialized species may thrive under conditions that might be deemed impossible. Such a case is the already mentioned hygrophytic almost aquatic Trichomanes reniforme on the sun-baked rocks of Rangitoto. Here are a few more examples: Crassula moschata Forst., a coastal halophytic herb, is one of the pioneer plants in the heavily manured ground just abandoned by penguins on the Snares Island. Colobanthus muscoides Hook. f., an herbaceous dense cushion plant growing normally on coastal rocks, is another early-comer on the above manured ground, but as conditions become favourable for less manure-tolerating plants both are replaced, tussock moor or Olearia forest being the climax association. Metrosideros scandens Sol., a root-climbing woody forest-liane, grows in some places on rocks close to the sea. Griselinia lucida Forst. f., so far as I am aware always either an epiphyte or a rock-plant, can be cultivated with ease as an ordinary garden-shrub.

The presence of closely related species side by side in the same association has a strong bearing on the mutation question, for it is reasonable to suppose with Leavitt (1907, pp. 210–12) that if natural selection, or even epharmony, is responsible for species-making, only one type will be present. As Leavitt writes, “Mutation breaks the species, and momentarily at least must give a polytypic aspect to the group within a specific

[Footnote] * H. H. Travers (1869) was of opinion that this tree was a very recent arrival, especially as an old resident, Mr. Hunt, did not know it, and as he found a seed on the shore of Pitt Island. I have given my reasons for be [ unclear: ] ving it an ancient constituent of the flora (1902, pp. 270–71), and have seen no reason to change my opinion.

[Footnote] † The case may not be as strong as it appears, since the seedlings are shade-loving, whereas those of N. fusca can tolerate far stronger light. There is also a fine tree in the dry Christchurch Domain, where the climate is much more unsuitable for indigenous forest-plants than Tapanui.

– 40 –

area” (loc. cit., p. 211). I cannot go fully into this important matter, but the following are rather striking examples. Many would not consider some of these plants “species,” they are so close; but so long as they are distinct entities which reproduce themselves “true “they meet the case as well or better.

Dracophyllum scoparium Hook. f., and another species considered by Cheeseman a form of this species (1909, p. 420) but by Kirk a var. of D. Urvilleanum, grow in the scrub of Campbell Islands. Celmisia vernicosa Hook. f. and C. campbellensis Chapman, a very rare plant, grow side by side in Auckland and Campbell Islands. Cotula Traillii T. Kirk, C. pulchella T. Kirk, and C. (obscura T. Kirk) ? grow together on coastal moor near Foveaux Strait. Two “species “of Acaena grow side by side on dunes in Southland: the one has more or less erect branches and longpeduncled flowers—it may be a var. of A. microphylla Hook. f.; the other is pressed most tightly to the ground, and has almost sessile flowers—it is A. microphylla var. pauciglochidiata Bitter. Both forms keep their distinctive characters for years when grown in garden-soil; intermediate forms occur amongst the wild plants which may be variants, mutants, or hybrids. Cotula atrata Hook. f. and C. Dendyi Cockayne sp. ined. occur on the same shingle-slip. Several absolutely distinct forms of Veronica buxifolia Benth. grow on the same subalpine herb-field (see Plate II, fig. 1). Rubus parvus Buch. and R. Barkeri Cockayne are in near proximity in the neighbourhood of Lake Brunner, Westland. Nothofagus fusca Oerst. and N. apiculata Cockayne grow in company in the forests at Day's Bay (Wellington) and Kaikoura (Marlborough). Astelia linearis Hook. f. and A. subulata Cheesem. grow side by side on mountain-moors in Auckland and Stewart Islands. Raoulia australis Hook. f. and R. lutescens Cockayne grow side by side on river-beds of the South Island Olearia Colensoi Hook. f. and O. Traillii T. Kirk grow mixed together in coastal scrub in Stewart Island. Cassinia albida Cockayne, C. Vauvilliersii Hook. f., C. fulvida Hook. f., and other closely related intermediate forms grow mixed on Mount Fyffe, Seaward Kaikoura Mountains. Two distinct forms of Cassinia Vauvilliersii grow just above the forest-line in Auckland Island (see Cockayne 1909A, p. 216). Cotula lanata Hook. f., C. propinqua Hook. f., and C. plumosa Hook. f. grow side by side on the shore of Auckland and Campbell Islands. Olearia ilicifolia Hook. f. and O. mollis Cockayne grow together in subalpine forest of Westland. Related Epilobia grow side by side in many places; some I know come true from seed. Poa foliosa Hook. f. and P. Tennantiana grow close together in Auckland Island. Celmisia sessiliflora Hook. f. and C. argentea T. Kirk grow side by side on certain alpine moors of the southern botanical province. Nothopanax simplex Seem. and N. parvum Cockayne are companion plants in the forest of Stewart Island and Westland. Carmichaelia Monroi Hook. f. and a related but more robust species not yet described* grow side by side on steppe and river-bed of the Canterbury Plain and eastern Southern Alps. Coprosma Petriei Cheesem. has two forms, one with claret-coloured drupes, and the other with faintly blue drupes; they grow side by side on montane steppe in the South Island. Coprosma Colensoi Hook. f. and C. Banksii Petrie occur side by side in many forests. Ranunculus Lyallii Hook. f. and

[Footnote] * What I take to be this plant received the herbarium name of C. humilis from D. Petrie many years ago. It has also been in cultivation along with C. Monroi Hook. f. in the Christchurch Domain for a long period.

– 41 –

a plant I considered R. Traversii, but which Cheeseman is of opinion is either a hybrid* on a new species, grow together on the Snowcup Mountains, Canterbury. Ranunculus Buehanani Hook. f. and the closely related R. Matthewsii Cheesem. grow in company on certain alpine herb- or fell-fields of western Otago. Anisotome pilifera Cockayne and Laing and its var. pinnatifidum T. Kirk grow in company on peat-covered rocks, &c., in the Southern Alps. Leptospermum ericoides A. Rich, and L. lineatum Cockayne grow together on northern dunes. Coriaria angustissima Hook. f., C. thymifolia Hunt. & Bonpl., and C. ruscifolia L. grow in proximity on Westland river-beds. Aciphylla Colensoi Hook. f. var. conspicua T. Kirk and the var. maxima grow near one another on certain herb-fields or in scrub on the Southern Alps. Two forms of Ourisia sessiliflora Hook. f., the one densely villous and with large flowers,* the other a smaller plant in all its parts, the leaves darker green and less hairy and the flowers fewer and smaller, occur on the same herb-field in the Southern Alps. Pittosporum rigidum Hook. f. and P. divaricatum Cockayne (see Plate I) occur in the same forest-area on the volcanic plateau. Sophora microphylla Ait. and S. prostrata Buchanan grow side by side in the bed of the River Waimakariri at the lower gorge. Doubtless a number of other examples could be found. The coupled plants are in all cases so closely related that they are considered by most New Zealand botanists either varieties of one species, the type and a variety, or forms not worthy of or that have not yet received a name. They are quite sufficient in number to show that it is not unusual for closely related hereditary plant entities to exist side by side for considerable periods.

The occurrence of distinct races of the same species at different points of its area of distribution is known in a few cases. As Leavitt says, such cases do not look like the work of mutation, nor can they be readily correlated with epharmony. The following are two striking examples: Rubus australis Forst. f. is a common plant both in forests and the open throughout the North, South, and Stewart Islands. In the northern part of the North Island it has, as a rule, much narrower leaves than in the southern part of its range—so much so that typical plants from the two areas have a very different appearance. The primary seedling-leaves seem to be identical in both forms: these are ovate or ovate-lanceolate, and coarsely toothed; they are soon succeeded by narrow leaves, much resembling those of R. parvus Buchanan, even as to their yellowish or slightly brownish marking. Seedlings growing in the forest-shade, and only 2.5-5 cm. tall, bear these narrow juvenile leaves, thus showing the form to be inherited, and not merely an epharmonic sun form. Since heredity is undoubted, the northern form demands a name. Styphelia fasciculata (Forst. f.), a heath-like small or tall shrub, extends from the North Cape to Canterbury and Westland. The adult form varies but little throughout its range, but the juvenile of the Auckland district has altogether broader leaves than that of the south. An example of a more local character is that of the mountain-herb Celmisia coriacea Hook. f., which from Mount Maungatua and other mountains in that part of Otago can be distinguished at a glance as a cultivated plant from other specimens collected on the

[Footnote] * The occurrence of this plant on Walker's Pass far from R. Monroi Hook. f. removes the suspicion of a hybrid origin, which Cheeseman adopted, partly at my own suggestion in the first instance.

– 42 –

actual dividing-range. The lowland form of the plant growing near the sea-cliffs at Charleston, west Nelson, is also distinct in appearance.

2. Isolation.

This special form of distribution is considered by some to be of the greatest evolutionary importance. The New Zealand biological area offers many ideal localities for geographical isolation, differing in degree, and it is interesting to see as to how far they afford examples of related species which appear to have either diverged recently from a stem form, or one of them to be the actual parent plant.

(a.) The Kermadec Islands.

The total number of species of vascular plants is 114, of which twelve are endemic.* These latter, one excepted, are closely related to, and in some instances almost identical with, New Zealand, Polynesian, or Norfolk Island plants.

(b.) The Three Kings Islands.

There is strong geological evidence that at no distant date these islands were united to the North Island. The total number of species of vascular plants is 143, of which five are endemic; with these Alectryon excelsum Gaertn. var. grandis Cheesem. may be included. Coprosma macrocarpa Cheesem., one of the five, is related to C. grandifolia Hook. f., and more distantly to C. robusta, both of which are present on the island. Pittosporum Fairchildii Cheesem. is allied by P. crassifolium A. Cunn. and P. umbellatum Banks & Sol. Veronica insularis Cheesem. is related to V. diosmaefolia R. Cunn., a species of the neighbouring mainland, and Paratrophis Smithii Cheesem. to P. opaca Brit. & Rend., while the fern Davallia Tasmani Field is not allied to any New Zealand species. None of the endemic plants, then, except the Alectryon, are particularly close to their mainland allies.

(c.) The North Cape.

This high promontory was undoubtedly quite recently an island. There are three endemic plants—Halorrhagis cartilaginea Cheesem. (a near relative of H. erecta Schind.), Geniostoma ligustrifolium A. Cunn. var. crassum Cheesem., and Cassinia amoena Cheesem. (probably related to C. Vauvilliersii Hook. f., but which latter is not found nearer than the volcanic plateau).

(d.) Islands lying to the Eastward of Auckland.

Veronica Bollonsii Cockayne, a species closely related to V. macroura Hook. f., is endemic on the Poor Knights Islands. Pittosporum intermedium T. Kirk, intermediate between P. tenuifolium Banks & Sol. and P. ellipticum T. Kirk, is found only on Kawau Island; only one plant has been found, and this has been destroyed.§

(e.) The Chatham Islands.

The total number of species plus named varieties is 236, of which thirty-one are endemic. The genera Myosotidium and Coxella are endemic and monotypic. The following is a list of the endemic plants; those related

[Footnote] * See Oliver, 1910, p. 150.

[Footnote] † See Cheeseman, 1891, pp. 419, 420.

[Footnote] ‡ Were not Mr. Cheeseman extremely cautious regarding the “creation “of species, &c., I should suspect this to be simply an unstable xerophytic form not very different from that with thick leaves common on the lava of Rangitoto Island.

[Footnote] § Cheeseman might consider this a hybrid were it not that P. ellipticum is not known either in Kawau or the neighbourhood (1906, p. 54).

– 43 –

more or less closely to New Zealand species are marked with an asterisk: Adiantum affine Willd. var. chathamicum Field (Filic.), *Poa chathamica Petrie, Festuca Coxii Hack. (Gram.), *Carex appressa R. Br. var. sectoides Kükenth., *Phormium tenax Forst. var. with broad thin drooping leaves (Liliac.), Geranium Traversii Hook. f. and var. elegans Cockayne (Geran.), *Linum monogynum Forst. f. var. chathamicum Cockayne (Linac.), *Plagianthus betulinus A. Cunn. var. chathamicus Cockayne (Malvac.), Aciphylla Traversii Hook. f., Coxella Dieffenbachii Cheesem. (Umbel.), *Corokia macrocarpa T. Kirk (Cornac.), *Styphelia robusta (Hook. f.), *Dracophyllum arboreum Cockayne, *D. paludosum Cockayne (Epacrid.), *Suttonia Coxii Cockayne (Myrsinac.), *Gentiana chathamica Cheesem. (Gentian.), Veronica Dieffenbachii Benth., V. Barkeri Cockayne, V. Dorrien-Smithii Cockayne, V. chathamica Buch., *V. gigantea Cockayne (Scroph.), *Coprosma chathamica Cockayne (Rubiac.), *Olearia semidentata Done., *O. chathamica T. Kirk, O. Traversii Hook. f., *Cotula Muelleri T. Kirk, C. Featherstonii F. Muell., *Senecio radiolatus F. Muell., *S. Huntii F. Muell., *Sonchus grandifolius T. Kirk (Compos.).

The nineteen “species “marked with an asterisk are closely related to forms found elsewhere in New Zealand, while sixteen of these are very close indeed. Veronica gigantea would certainly be considered a variety of V. salicifolia Forst. were it not for its distinct juvenile form, which still persists up to a stature of at least 80cm., and its arboreal habit. It is the only true forest-veronica, and it may be that the juvenile form is a direct adaptation to forest-undergrowth conditions.

(f.) Stewart Island.

A number of species have, as yet, been collected only on Stewart Island, but in the face of the fact that year by year shows more of the plants thought to be endemic fairly common on the mainland, &c., it is quite possible that the island contains no endemic species.

(g.) The Subantarctic Islands of New Zealand.

There are 195 species and named varieties, of which fifty-one a [ unclear: ] endemic, nineteen of these being closely related to New Zealand species. No list is given here, as these endemic species are treated of by Cheeseman with considerable detail (1909, pp. 463–66). With regard to special endemism in the various groups, the Auckland Islands have six species, the Campbells four, Antipodes Island four, Macquarie Island three, and the Snares two.

(h.) Isolation on the Main Islands.

Endemism is not confined to isolated islands, but the various floral districts contain their peculiar species and forms. The most striking examples are western Nelson and western Otago, with thirty-three and thirty-eight endemic species respectively. The northern part of Auckland (thirteen species), Marlborough (fourteen species), and other localities show a distinct local endemism. It is obvious, then, that a strong endemism can exist apart from such a barrier as a wide stretch of ocean. But figures such as the above are not final; further investigations may decrease or even increase them. Also, it is certain that not all the species included have originated in the “isolated “areas; some of the most distinct have probably been much more widely spread, and are “relics “merely.

The continuity of distribution of species of the New Zealand flora varies from those with a fairly continuous distribution to those which occur in only a few localities far distant from one another. Notable examples of extreme discontinuity are: Danthonia antarctica Hook. f., common in

– 44 –

Auckland and Campbell Islands, but confined elsewhere to a few rocky points and small islands in the far north of the North Island; Urtica australis Hook. f., common in Chatham, Antipodes, and Auckland Islands, but in New Zealand proper occurring only on Dog and Centre Islands, Foveaux Strait; Drosera pygmaea D.C., only recorded from near Kaitaia in the extreme north and the Bluff Hill in the extreme south; Pittosporum obcordatum Raoul, occurs sparingly near Kaitaia, and Akaroa, Banks Peninsula; Plagianthus cymosus T. Kirk, only recorded from Dunedin, Lyttelton, some of the Marlborough Sounds, and Kaitaia; Suttonia chathamica Mez, common in the Chatham Islands, and found in two localities in Stewart Island; Lepyrodia Traversii F. Muell., common in Chatham Island, and found in certain bogs of the Waikato and at one locality near Kaitaia; Styphelia Richei Labill., common in Chatham Island, and found elsewhere only near the North Cape; Melicytus macrophyllus A. Cunn., common in certain Auckland forests, but absent elsewhere, except one locality near Dunedin Other examples of discontinuous distribution, though more connected than the above, include Elaeocharis sphacelata R. Br., Dracophyllum latifolium A. Cunn., Clematis afoliata Buch., Quintinia acutifolia T. Kirk, Celmisia Traversii Hook. f., Pseudopanax ferox T. Kirk, Carmichaelia gracilis J. B. Armstg., Coprosma rubra Petrie, Veronica speciosa R. Cunn., &c. Were there merely one or two cases the discontinuous distribution might be attributed to chance, but as there are numerous cases, and as these gradually merge into examples of greater and greater continuity, it is probable that the species in most cases were at one time more widely spread, and that in the extreme cases as above we are face to face with the phenomenon of a species naturally on the verge of extinction.

IX. Evolution In The Genus Veronica In New Zealand.

The New Zealand flora, as already pointed out, possesses many genera containing very “variable species,” which are of much interest for evolutionary studies. Of all such, Veronica is the most instructive, illustrating, as it does, the general principles of evolution apart from any theories as to method.

Cheeseman admits eighty-four species, but the view he takes is a most conservative one, and probably without forsaking the ideals of orthodox taxonomy some thirty more species could be conveniently added to the list. Were, however, that school of botany which is dealing with Rosa, Rubus, Hieracium, and Crataegus in the Northern Hemisphere to study the New Zealand forms, several hundreds of species would be forthwith “created.” Should this ever be done without experimental culture of each proposed form the work will be biologically useless.

The species differ both epharmonically and floristically. The former concerns distinctions between groups of forms rather than between species, while the latter treats of the specific marks.

There are two main classes—the shrubby and the herbaceous—together with the suffruticose. The multitude of forms, with but few exceptions, are connected, and a great number more or less intergrade in a linear series. There is every evidence, then, of descent from a common ancestor, which, considering the genus beyond New Zealand as well as within its confines, would probably be an herbaceous plant with a didymous capsule such as V. Chamaedrys L. Further, the plasticity of many “species “and the astonishing variability suggest that changes of form are, biologically speaking, in rapid progress at the present time.

– 45 –

The New Zealand species, with but few exceptions, reproduce themselves readily and rapidly from seed, can be easily grown from cuttings, and are not restricted to any special soil. Some respond quickly to change of environment. The genus occurs in all parts of the New Zealand region, except Antipodes and Macquarie Islands. It has representatives in almost every plant formation, but there is only one true forest species (V. gigantea, of Chatham Island). An analysis of distribution shows that seventeen species are coastal, thirteen do not ascend beyond 300 m. altitude, ten to between 300m. and 900m., thirteen to 900m. and less than 1,200 m., and forty-three to that altitude and upwards, while fifty-two of the ninety-six may be considered strictly mountain species.

Regarding their growth-forms, perhaps six species might be considered herbaceous; the remainder are all more or less woody, the great majority being shrubs. Beyond New Zealand there is one shrubby Veronica in Fuegia and the Falkland Islands, V. elliptica Forst. f., identical with or closely related to one or other of that series of forms known as V. elliptica in New Zealand, and V. formosa R. Br. and V. densifolia F. Muell. of Tasmania and south-east Australia respectively.

Leaving the herbaceous cushion plants, formerly referred to the genus Pygmaea, on one side for the present, the remainder of the herbaceous and suffruticose veronicas (Division Euveronica J. B. Armstg.) are distinguished by their didymous capsule. But the shrubby V. loganioides J. B. Armstg. has a similar capsule. This plant resembles in many respects a juvenile form of the whipcord section of Division I, Hebe. There is another epharmonically similar plant, V. cassinioides Hort., which has a capsule of the Hebe type, and which represents a fixed juvenile form of a whipcord Veronica, such as that fixed or semi-fixed form of V. tetragona Hook., which occurs occasionally on the volcanic plateau (see Plate V, fig. 2). Still more is the relation to whipcord veronicas shown in the toothed leaves of reversion shoots. With a broadening of leaf, a not uncommon occurrence, there is a close approach to V. buxifolia Benth. In considering the phylogeny of the species of Veronica the change from herb to shrub would be epharmonic, as may now be seen in the series of forms from just suffruticose to almost shrubs. In such manner V. loganioides might arise, and, the form of capsule changing by mutation, there would be V. cassinioides, which on the one hand could develop by way of V. buxifolia into the mesophytic species, or through pressing of leaves to the stem, and a certain amount of reduction, into the xerophytic whipcord forms. Of course, I do not imagine these are the actual ancestral species, but it does not seem absurd to take them as approximative to such. Some further details may shed a little light on the matter.

The shrubby veronicas fall into three epharmonic classes, using Cheeseman's synopsis. The first would include from V. speciosa R. Cunn. to V. pimeleoides Hook. f., the second from V. Gilliesiana T. Kirk to V. uniflora T. Kirk, and the third from V. macrantha Hook. f. to V. Raoulii Hook. f. The first class shows a leaf gradually decreasing in size, and varying from the willow form, broader or narrower as the case may be, to the small more or less oblong or ovate leaf of so many of the subalpine species—that is, there is a reduction of leaf-surface in accordance with increase of xerophytic conditions. Where lowland species occupy xerophytic stations large leaves are thickened in texture, as in V. Dieffenbachii Benth., V. speciosa R. Cunn., and V. macroura Hook. f., all plants of coastal rocks; or reduced and thickened, as in V. chathamica Buch., another coastal-rock plant; or much

– 46 –

reduced in size, as in V. diosmaefolia R. Cunn., a heath-plant—indeed, there are few species whose leaf-form cannot be referred to evident epharmony.

The general habit of the species is often strikingly epharmonic. In point of fact, all branch on the same plan, but density or looseness of branching in its extremes makes very different plants, as in the far-spreading, open, and stragglingly branched V. Cookianum Col. and V. Dieffenbachii Benth., and the close ball-like V. buxifolia var. odora T. Kirk, V. Traversii Hook. f., and many of the subalpine semi-xerophytic species. Still more xerophytic species have the prostrate form, as V. chathamica, a plant of wind-swept and spray-swept coastal rocks, and V. pinguifolia Hook. f., in some of its numerous forms, as it hugs dry alpine rocks or the stony surface of fell-field. It is instructive, too, to see how one and the same Linnean species varies in the growth-forms of its components. Thus V. buxifolia Benth. may be either a ball-like shrub, a low erect open little-branched shrub, or sparsely branched and prostrate. Its leaves, too, vary from patent to imbricating; while as for small leaf-variations, they are without end. The degree of imbricating of leaves is a striking epharmonic feature in these small-leaved veronicas, and Cheeseman uses it, but in a guarded manner, as an aid to identification. But the truth is, the individuals of a well-defined form vary much in this regard according to their surroundings, while there appears also to be non-epharmonic variation of this character.

A more xerophytic station in general than that of the subalpine species of class 1 is demanded by those of class 2. Here reduction of leaf and imbricating reach their maximum in the whipcord forms. These have fully developed though small leaves as seedlings and on reversion shoots, and are thus united to Veronica Gilliesiana, T. Kirk, Hook. f., and others whose leaves are not so much reduced. Classes 1 and 2, as here defined, seem to be connected by V. buxifolia Benth., as a study of its seedling form shows.* But this latter is also related to V. cassinioides Hort., which, as already shown, is a juvenile or ancestral whipcord Veronica which may be linked with suffruticose and herbaceous species by V. loganioides J. B. Armstg. The relation, then, if my supposition be accepted, between such a species as V. buxifolia or some form such as V. cassinioides is so close that favourable epharmonic conditions should convert the one into the other in course of time. The cupressoid growth-form of these whipcord veronicas may easily have appeared epharmonically several times. Each time there would be some slight difference in the form evoked, and thus some of the species of whipcord Veronica may have originated independently and not from one ancestral cupressoid form, and there may have been actual

[Footnote] * Details are given by me (1901, pp. 282–86) under the name V. odora Hook. f., which, however, is now known through the researches of Cheeseman (1909) to be distinct from the plant in question, which is V. buxifolia Benth. var. odora T. Kirk. PI 11 in the above paper should be consulted, as it shows the relation in form between the juvenile leaves of V. buxifolia var. odora and V. Armstrongii T. Kirk, a whipcord Veronica.

[Footnote] † Regarding polygenetic origins, Chilton wrote (1884, p. 156), “Suppose the marine ancestor of the terrestial Isopoda to be widely spread, and to inhabit the shores of, say, New Zealand and England, and that in each case certain animals began gradually to leave the sea and make their home on the land, at first keeping within the range of the spray, as Ligia still does, but afterwards leaving the sea altogether, would not the new conditions in which these animals would be placed, being practically the same in both countries, produce in each case the same effect, so that the variations which would be preserved would be the same in the two cases, and hence the animals, although arising independently from the same marine ancestor, might so far resemble one another as to be placed in the same genus or even in the same species ? “Guppy (1907) should also be consulted.

– 47 –

polygenetic development of species. This polygenetic origin of form, if not of species, is the more likely, as the form exists in other families, while the distribution of the species shows that, though some are widespread, there are a number of species of restricted distribution—e.g., V. Langii Cockayne (Stewart Island), V. Hectori Hook. f. (western Otago), V. propinqua Cheesem. (Mount Maungatua and some other Otago mountains), V. salicornioides Hook. f. (Nelson), V. Astoni (Tararua Mountains), V. tetragona (volcanic plateau), and others not yet described.

Veronica Haastii Hook. f., V. epacridea Hook. f., and V. Petriei T. Kirk are not definitely connected with the rest of class 2, and may be considered a side branch, with modified leaves.

Class 3 form a distinct line of descent to itself, and its connection with any other branch of the genus is not clear. Two species are moor-plants, and the remainder rock-plants; their growth-forms are epharmonic. The branched panicle of V. Hulkeana F. Muell., V. Lavaudiana Raoul, and V. Raoulii Hook. f. remove them from the rest of the class. Nevertheless branching of the inflorescence is merely a question of degree, and occurs at times in various species—e.g., V. Traversii Hook. f., where it is unexpected—while in others a similar inflorescence is a specific character (V. diosmaefolia, V. Menziesii Benth.).

Regarding the herbaceous species, V. pulvinaris Benth. & Hook. belonging to Pygmaea, their leaves are not arranged quadrifariously. By some they are regarded as forming a distinct section of the genus. At present it is impossible to assign them a place in the direct line of descent. They are cushion plants, and epharmonically similar to Myosotis pulvinaris Hook. f.

The suffruticose veronicas (V. catarractae Forst. f., V. Lyallii Hook. f., and V. Bidwillii Hook. f.) are closely related to one another—so closely, indeed, that it is hard to assign limits to any as a Linnean species, and the simplest method from that standpoint would be to unite all three.

X. Concluding Remarks.

The object of this paper is to supply material for consideration by students of evolution culled from a field which, although not altogether neglected, is much less cultivated for the supply of evolutionary pabulum, especially by English writers, than is the wide domain of zoology, whence come the bulk of the facts of so many works on evolution.

Whatever of value there may be in this ecological material lies in the fact that it is drawn from an isolated and virgin vegetation, and one, too, where the grazing animal played a most insignificant part compared with its role in the Old World.

The details have not been selected to support any particular theory, though, of course, as ecological observations are the basis of the paper, the relation of plant to environment takes the leading place.

By one celebrated school of biologists the ultimate inheritance of characters* evoked by stimuli affecting the body-cells is either considered impossible or an occurrence so rare as to be negligible, while such evidence as I have advanced is looked upon as worthless, or, at best, as quite

[Footnote] * For years Henslow has battled strenuously for the cause of the inheritance of acquired characters, but without receiving the attention his works deserve; in fact, many writers seem acquainted only with his “Origin of Floral Structures,” and neglect altogether his much more convincing “Origin of Plant Structures,” a work full of suggestive material.

– 48 –

insufficient. But another and equally famous school believe such inheritance to be a more or less frequent occurrence, botanists, as a rule, being more in its favour than are zoologists.

Speaking of theories of evolution generally, there seems, good reason to consider that such, if not premature, are chiefly of value as a stimulus to biological research. Our ignorance as to the minute structure, the chemistry, and the physiology of the protoplasm is profound. Nothing is known as yet regarding the actual cause of variation. An epharmonic stimulus could do nothing were it not that the inner constitution of the plant is already able to respond—i.e., the “machinery “is there ready to produce the possibly epharmonic variation so soon as it gets the necessary touch.

The construction of elaborate theories is not the method by which progress can be made. Actual experiments in the garden, the laboratory, and the field can alone lead to the truth. Even in taxonomy, only experiment can actually decide as to stable and hereditary forms. But observations from nature are also demanded, and here ecology comes in, with the attempt to make use of the wild-plant world, where there are species in the making, as a source of observation. The duty of the ecologist is the collecting of facts in as accurate a manner as possible. The study of epharmony in its manifold phases is urgently required. Its vigorous prosecution should yield a rich harvest of observations, to be examined in the light of experimental evolution.

XI. Literature Cited.*

Armstrong, J. B. 1881. “A Synopsis of the New Zealand Species of Veronica Linn., with Notes on New Species.” Trans. N.Z. Inst., vol. 13, p. 344.

Balfour, I. B. 1879. “The Collections from Rodriquez—Botany.” Phil. Trans. R.S., vol. 168, p. 302.

Bitter, G. 1911. “Die Gattung Acaena.” Stuttgart.

Blaringhem, L. 1907. “Mutation et Traumatisme.” Paris.

Buchanan, J. 1870. Introductory Remarks to “List of Plants found in the Northern District of the Province of Auckland.” Trans. N.Z. Inst., vol. 2, p. 239.

—— 1871. “On some New Species and Varieties of New Zealand Plants.” Trans. N.Z. Inst., vol. 3, p. 208.

Burns, G. P. 1911. “Edaphic Conditions in Peat Bogs of Southern Michigan.” Bot. Gaz., vol. 52, p. 105.

Cheeseman, T. F. 1891. “Further Notes on the Three Kings Islands.” Trans. N.Z. Inst., vol. 23, p. 408.

—— 1906. “Manual of the New Zealand Flora.” Wellington.

—— 1907. “Contributions to a Fuller Knowledge of the Flora of New Zealand.” Trans. N.Z. Inst., vol. 39, p. 439.

—— 1908. Ibid., No. 2. Trans. N.Z. Inst., vol. 40, p. 270.

—— 1909. “On the Systematic Botany of the Islands to the South of New Zealand.” The Subant. Islands of N.Z., vol. 2, p. 389.

Chilton, C. 1884. “The Distribution of Terrestial Crustacea.” N.Z., Journ. Sci., vol. 2, p. 154.

Clements, F. E. 1905. “Research Methods in Ecology.” Nebraska.

[Footnote] * Works consulted but not referred to in the text are not included, except in a few instances.

– 49 –

Cockayne, L. 1901. “An Inquiry into the Seedling Forms of New Zealand Phanerogams and their Development, Part IV.” Trans. N.Z. Inst., vol. 33, p. 265.

—— 1902. “A Short Account of the Plant Covering of Chatham Island.” Trans. N.Z. Inst., vol. 34, p. 243.

—— 1904. “A Botanical Excursion during Midwinter to the Southern Islands of New Zealand.” Trans. N.Z. Inst., vol. 36, p. 225.

—— 1907. “Note on the Behaviour in Cultivation of a Chatham Island Form of Coprosma propinqua.” Trans. N.Z. Inst., vol. 39, p. 378.

—— 1907A. “On the Sudden Appearance of a New Character in an Individual of Leptospermum scoparium.” New Phytol., vol. 6, p. 43.

—— 1908. “Report on a Botanical Survey of the Waipoua Kauri Forest.” Wellington.

—— 1909. “Report on a Botanical Survey of Stewart Island.” Wellington.

—— 1909A. The Ecological Botany of the Subantarctic Islands of New Zealand.” The Subant. Islands of N.Z., vol. 1, p. 182.

—— 1910. “On a Non-flowering New Zealand Species of Rubus.” Trans. N.Z. Inst., vol. 42, p. 325.

—— 1911. “On the Peopling by Plants of the Subalpine River-bed of the Rakaia (Southern Alps of New Zealand).” Trans. Bot. Soc. Edinb., vol. 24, p. 104.

Cook, O. F. 1907. “Aspects of Kinetic Evolution.” Proc. Wash. Acad. Sci., vol. 8, p. 197.

Costantin, J. 1898. “Lea Vegetaux et les Milieux Cosmiques.” Paris.

Cross, B. D. 1910. “Observations on some New Zealand Halophytes.” Trans. N.Z. Inst., vol. 42, p. 545.

Darwin, C. 1899. “The Origin of Species.” London. (6th ed.) — 1905. “The Variation of Animals and Plants under Domestication.” (Popular edition, edited by Francis Darwin.)

—— 1908. Presidential Address. Rep. Brit. Assoc.

Dendy, A. 1902. “The Chatham Islands: a Study in Biology.” Mem. and Proc. Manch. Lit. and Phil. Soc., vol. 46, pt. 5.

—— 1903. “The Nature of Heredity.” Rep. S.A.f. A.A.S., vol. 1.

Diels, L. 1906. “Jugendformen und Blütenreife im Pflanzenreich.” Berlin.

Goebel, K. 1889–93. “Pflanzenbiologische Schilderungen.” Marburg.

—— 1900–5. “Organography of Plants.” Oxford.

—— 1908. “Einleitung in die Experimentelle der Pflanzen.” Leipzig and Berlin.

Griffen, E. M. 1908. “The Development of some New Zealand Conifer Leaves with Regard to Transfusion Tissue and to Adaptation to Environment.” Trans. N.Z. Inst., vol. 40, p. 43. 1907. “Plant-distribution from an Old Standpoint.” Author's copy of paper read before the Vict. Inst

Hall, H. M. 1910. “Studies in Ornamental Trees and Shrubs.” Univ. of Cal. Pub. in Bot., vol. 4, p. 1.

Haswell, W. A. 1891. “Recent Biological Theories.' Rep. A.A.A.S., vol. 3, p. 173.

Henslow, G. 1895. “The Origin of Plant Structures.” London.

—— 1908. “The Heredity of Acquired Characters in Plants.” London.

– 50 –

Hooker, J. D. 1853. “Flora Novae-Zelandiae.” Vol. 1. Introductory Essay, p. i.

Kirk, T. 1871. “On the Botany of the Northern Part of the Province of Auckland.” Trans. N.Z. Inst., vol. 3, p. 166.

—— 1889. “The Forest Flora of New Zealand.” Wellington.

—— 1896. “The Displacement of Species in New Zealand.” Trans-N.Z. Inst., vol. 28, p. 1.

Klebs, G. 1903. “Willkürliche Entwickelungsänderungen bei Pflanzen.” Jena.

—— 1910. “Influence of Environment on the Forms of Plants.” Darwin and Modern Science, p. 223.

Leavitt, G. G. 1907. “The Geographic Distribution of Closely Related Species.” Am. Nat., vol. 41, p. 207.

MacDougal, D. T. 1911. “Inheritance of Habitat Effects in Plants.” Plant World, vol. 14, p. 53.

Massart, J. 1910. “Esquisse de la Géographie botanique de la Belgique.” Bruxelles.

Oliver, R. B. 1910. “The Vegetation of the Kermadec Islands.” Trans. N.Z. Inst., vol. 42, p. 118.

Romanes, G. J. 1893–97. “Darwin and After Darwin.” London.

Scott-Elliott, G. F. 1910. “The Waning of Weismannism.” Journ. R. Hort. Soc., vol. 35, p. 327.

Speight, R. 1911. “The Post-glacial Climate of Canterbury.” Trans. N.Z. Inst., vol. 43, p. 408.

Thomson, G. M. 1901. “Plant-acclimatization in New Zealand.” Trans. N.Z. Inst., vol. 33, p. 313. (Contains various supplementary notes by D. Petrie.)

Travers, H. H. 1869. “On the Chatham Islands.” Trans. N.Z. Inst., vol. 1, p. 173.

Travers, W. T. L. 1870. “On the Changes effected in the Natural Features of a New Country by the Introduction of Civilized Races.” Trans. N.Z. Inst., vol. 2, p. 299.

Vries, H. de. 1901–3. “Die Mutationstheorie.” Leipzig.

—— 1905. “Species and Varieties, their Origin by Mutation.” Chicago.

Wallace, A. R. 1889. “Darwinism.” London.

Warming, E. 1909. “Oecology of Plants.” Oxford.

Weismann, A. 1910. “The Selection Theory.” Darwin and Modern Science, p. 18.

Williams, W. L. 1904. ‘Abnormal Growth of a Plant of Phormium Colensoi.’ Trans. N.Z. Inst., vol. 36, p. 333, and pl. 25.