Go to National Library of New Zealand Te Puna Mātauranga o Aotearoa
Volume 42, 1909
This text is also available in PDF
(552 KB) Opens in new window
– 348 –

Art. XLIV.—Botanical Evidence against the Recent Glaciation of New Zealand.

[Read before the Otago Institute, 22nd September, 1909.]

In his presidential address recently (11th May, 1909) delivered before our Institute, Professor Park stated that “during the progress of his geological work he had been fortunate enough to obtain evidence to show that New Zealand at one time had a glacial period similar to that of the Northern Hemisphere. He had been able to prove that Otago and the greater part of the South Island were covered with a continuous ice-sheet, probably an extension of the South Polar ice.” This glaciation occurred, according to the President, in the Pleistocene period.

It seemed to me, at first sight, impossible to explain many facts in connection with the existing flora of New Zealand in face of such a pronouncement, and I have accordingly sought to test the accuracy of this theory from the botanical standpoint. Such evidence as is available seems to me to be quite opposed to such a theory.

Professor Park's claim means that the greater part of the South Island (which, according to him, was only separated from the North Island in the Pleistocene period) was at that time covered with an ice-sheet (exceeding 7,400 ft. in thickness in the Wakatipu basin) synchronous with that which covered the Northern Hemisphere. To quote his own words as to the effects of such an ice-sheet, “The forests were overwhelmed and completely obliterated, the animals and men inhabiting … areas retreated before the terrible blighting wall of ice. … The evidence that this mass of ice had been all over Central Otago was so fresh that they might easily imagine it had retreated only yesterday.”

There are three lines of botanical evidence which appear to me to disprove a recent glacial epoch, as distinct from merely increased glaciation due to elevation in the central regions. They are—(I) the occurrence in the South Island and in the antarctic islands lying to the south of a remarkably specialised group of genera and species, some of which range into or are allied to forms occurring in Australia and elsewhere, but not

– 349 –

found in the North Island; (II) the absence of deciduous-leaved plants in New Zealand; and (III) the occurrence of a xerophytic or desert flora. All these conditions demand an amount of time for their evolution much in excess of that granted by any theory of Pleistocene glaciation.


According to Cheeseman's Flora, there are 332 genera and 1,415 species of flowering-plants in New Zealand. Of this number, two genera and forty-five species are peculiar to the antarctic islands (Macquarie, Campbell, Auckland, Antipodes Islands, and the Snares); while twenty-six genera and 452 species are found in the South Island which do not occur in the North Island.

If we take the genera in detail, we find the two genera peculiar to the antarctic islands to be more or less allied to others in the South Island. Thus Stilbocarpa (Araliaceœ), found in most of the islands, is allied to and probably derived from Aralia, a genus found in the Snares and Stewart Island, and in the south-west corner of the South Island. Of the three species of Pleurophyllum, P. speciosum and P. criniferum occur in the Auckland and Campbell Islands, while P. Hookeri is also found in these groups and in Macquarie Island. The genus is closely allied to Celmisia, “from which it is separated rather by the very distinct and peculiar habit than by any structural characters of importance.”

But when we examine the genera (nine of which are peculiar) found in the South Island and not in the North, we meet with such diversity of structure and form that we are compelled to postulate long periods of time to account for their great differentiation, or else our views of evolutionary development are all wrong.

Among the Cruciferœ the genus Pachycladon, represented by one species found only in the mountain regions of Central and West Otago, is quite distinct from all other forms in New Zealand, and, indeed, is somewhat peculiar in its characters, being “intermediate between the tribes Sisymbrieœ and Lepidineœ.” Another remarkable genus—Notothlaspi—has two species singularly developed to resist drought. This genus also is quite unlike any other. The genus Hectorella is not closely allied to any other genus of the order Portulaceœ, to which it belongs. Corallospartium crassicaule, the only representative of a remarkable genus of Leguminosœ, is allied to Carmichaelia, itself a genus peculiar to New Zealand (and Lord Howe Island). Another peculiar type of this order is Notospartium, which, though externally like Carmichaelia, differs in various points, such as its many-jointed pod, a sort of character which does not probably arise by gradual development from some other type.

Among Compositœ the only genus confined to the South Island is Haastia, a most remarkable group of four species of woolly-leaved plants. Here we have not only a distinct genus, but four distinct species developed since the separation of the Islands.

Tetrachondra is a genus of one species (T. Hamiltoni), which has hitherto only been found in the east and south of Otago. It is so anomalous in its characters that systematists are not even agreed as to its position. Kirk placed it in the genus Tillœa (nat. ord. Crassulaceœ); Oliver transferred it to the Boragineœ, among which Cheeseman also places it; while Hallier suggests that it should be considered an anomalous member of the Scrophulariaceœ, and that its nearest ally is the section Pygmaea of Veronica.

– 350 –

Anagosperma is a monotypic genus of the Scrophularineœ, allied to but distinct from Euphrasia. (Siphonidium of Armstrong is probably the same, according to Cheeseman.) Lastly, Townsonia is a monotypic orchid which has hitherto been found only in Nelson and Westland.

Making all allowance for differences of opinion as to what constitutes a generic distinction, yet the mere occurrence of such a group of specialised forms as these nine genera represent in the South Island of New Zealand alone surely implies a fairly long ancestry, and it seems quite unlikely that all these plants have spread from the North Island since its separation (if this took place in Pleistocene times), and have had time to differentiate so widely from other types.

But, leaving generic forms which are new, we find even more striking evidence of specific distinctions which are highly interesting.

(1.) There exist in the antarctic islands two species which occur in other parts of the world but not in the islands to the immediate north. If these islands were buried under an ice-sheet, all the vegetation must have been destroyed. In what way, then, could these plants have got to their present habitats within very recent times ? They are Azorella Selago, found in Macquarie Island, and also in Kerguelen, the Crozets, Marion and Heard Islands, and Fuegia; and Cotula plumosa, found in all the antarctic islands, and also in Kerguelen and the Crozets.

(2.) Among South Island species not found in the North Island, a number occur whose allies are found in other parts. I have specified twenty-four species, as follows: Swainsonia novœ-zealandiœ, the only species in New Zealand, but belonging to a genus common in Australia; Aralia Lyallii, also unique, but allied to species in America and in the Northern Hemisphere; Actinotus novœ-zealandiœ, with a distribution similar to Swainsonia; Crepis novœ-zealandiœ, an anomalous form, most of the other species of the genus occurring in the Northern Hemisphere; Isotoma fluviatilis, similar to Swainsonia; Pernettya nana, a heath almost identical with a Tasmanian species, while all the other species are American; Scutellaria novœ-zealandiœ, with distribution similar to the last; Exocarpus Bidwillii, Poranthera microphylla, Lyperanthus antarcticus, Iphigenia novœ-zealandiœ—all four with immediate allies in Australia and other regions; Triglochin palustre, also found in South America and in the Northern Hemisphere; Centrolepis strigosa, also in Australia and Tasmania; Gaimardia setacea, whose nearest allies are in Tasmania and South America; Uncinia Sinclairii, also found in Fuegia; Uncinia tenella, also found in Australia; Carex appressa, C. lagopina, C. trifida, and C. flava, and four grasses (Stipa setacea, Agrostis magellanica, Trisetum subspicatum, and Koeleria Kurtzii), all with widespread distribution.

(3.) In addition to these species which have their nearest relations in other parts of the globe, or are themselves more or less widely distributed, there is a remarkable assemblage of species peculiar to the South Island or the more southern islands, and which do not range into the North Island. I need not give a full list of these, but must specify the most remarkable in order to show how greatly they are differentiated. They include Ranunculus Lyallii and R. Buchanani, with large white flowers, and the former with great peltate leaves; Acœna glabra, which has lost the spines and barbs on the calyx so characteristic of the genus; Olearia insignis, which differs from other species of the genus “in the large broadly ovoid involucre with the bracts in very many series, and in the pappus of perfectly equal hairs”; the singular group of large-flowered Olearias, which includes

– 351 –

O. Operina, O. angustifolia, O. Traillii, O. Colensoi, and O. Lyallii; Olearia fragrantissima, with its heads collected into aggregate or compound heads: no less than thirty-four species of Celmisia, showing a great degree of modification; the remarkable Raoulias of the “vegetable sheep” type (R. eximia and R. mammillaris); Raoulia grandiflora, a type differing from all the others of the genus; several highly differentiated species of Senecio, including the climbing S. sciadophilus, and S. geminatus, a form allied to certain genera found in Juan Fernandez, and so distinct that Hooke formerly placed it in a distinct genus, Traversia; Phyllachne clavigera, a most singular plant, allied to a Fuegian form; Lobelia Roughii, “a most distinct species, quite unlike any other” of the genus in its three-lipped corolla; Mitrasacme novœ-zealandiœ and M. montana, the latter either identical with or very nearly allied to a Tasmanian species, while the other species are mostly Australian or south Asiatic. Then we have the remarkable development of the genus Veronica, in which occur some curious groups of forms: (1) the “whip-cord” species—V. Gillesiana, V. tetrasticha, V. quadrifaria, V. tumida, V. lycopodioides, V. Hectori, V. coarctata, V. salicornioides, V. Armstrongii, V. propinjua, and V. cupressoides; (2) the three moss-like species—V. pulvinaris, V. Thomsoni, and V. ciliolata, originally separated by Hooker as a separate genus Pygmaea; (3) the anomalous V. loganioides, a species quite distinct from any other of the genus, and whose relationships are somewhat obscure; and (4) the curious dried-up-looking forms V. Cheesemanii and V. canescens. Altogether, the Veronicas constitute a group of forms which have probably been long isolated to bring about the remarkable differentiation which they now exhibit.

I have gone minutely into these details, and drawn my facts from an analysis of Mr. Cheeseman's Manual, because it is important in such a connection to show as fully as possible the evidence on which I base my opposition to the idea of a recent glacial epoch in New Zealand.


One of the effects of the recent glaciation of the Northern Hemisphere was to produce among the woody plants which followed the ice in its retreat northwards a deciduous-leaved character. Woody plants characteristic of a cold temperate region, and liable to meet with heavy snowfalls in the winter months, can only escape destruction by special adaptation to such conditions. The simplest way, apparently, is to become deciduous on the approach of the cold season. Now, if we compare the characters of the British and the New Zealand floras in this respect, we find an interesting dissimilarity, which perhaps is best shown by tabulating them as follows:—

Britain. Per Cent of the Whole. New Zealand. Per Cent. of the Whole.
Total number of species 1,223 1,415
Woody species 91 7.4 450 31.8
Evergreen trees and shrubs 26 28.5 398 88.3
Deciduous trees and shrubs 65 71.4 7 (partly) 1.5 (?)
Leafless, or nearly so 45 10.0
– 352 –

Here we find that the woody plants of Britain—a region glaciated within recent geological times—constitute only 7.4 per cent. of the whole flora. Of the ninety-one woody plants in Britain, excluding prostrate and creeping species, sixty-five, or 71.4 per cent., are quite deciduous. The remaining twenty-six species are evergreen, but their foliage is of such a character that they are not liable to take any damage from snow. I append a list of these forms,* from which it will be seen that some of them are small-leaved heaths and similar plants.

New Zealand has 1,415 species of plants, but 450 of these, or 31.8 per cent., are woody. Of this number, no less than 398, or 88.3 per cent., are evergreen; forty-five, or 10 per cent., are leafless or have only minute appressed leaves; while seven species, or less than 1.5 per cent., are more or less deciduous. These are Gaya Lyallii, which is deciduous only at high levels and in the south end of the South Island, but is evergreen in the valleys of Nelson and Westland; Aristotelia racemosa, A. Colensoi, and A. fruticosa, which are also deciduous only at high levels; Coriaria thymifolia, also tending to become deciduous at high levels; and Fuchsia excorticata and F. Colensoi, which are deciduous at least in the southern part of New Zealand. Thus in the whole of New Zealand there is no absolutely deciduous species; and this is, to my mind, a strong argument against recent glaciation.


A characteristic feature of the flora of New Zealand is the presence of a large xerophytic element, a very considerable group of the plants being specialised to endure desert conditions.

All green-leaved plants are furnished with stomata for the breathing in and out of the gases which they feed on and eliminate, and for the transpiration of water-vapour. Plants growing in peaty moorlands, and such as grow on the sea-coast, even in climates where there is relatively considerable humidity, are frequently specially modified to resist overtranspiration, and these modifications take the form of archings of the upper surfaces of the leaves, thickening of the tissues, and diminution of the number of stomata on all their breathing-surfaces. Eliminating these locally modified plants, which are common to many regions of the globe, we find in New Zealand, and especially on the eastern side of the Alps, and the hills and plains lying to the east of the dividing-range, an assemblage of plants—approximately estimated by me at 220 species—which are peculiarly adapted to withstand drought. Their breathing-surfaces are more or less covered with hairs, sometimes produced in extraordinary profusion, or they have greatly thickened epidermis with deep-set stomata, or the upper surfaces which have no stomata arch over and thus tend to close the pores on the sheltered undersides. In many the leaves become modified (e.g., many species of Veronica), so as to lie closely pressed against the stem, or the leaves are more or less completely aborted (Discaria toumatou and species of Carmichaelia), the breathing-pores in such cases being produced on the epidermis of the branches and stems.

[Footnote] * Berberis vulgaris, Ilex aquifolium, Euonymus europœus, Genista tinctoria, Ulex europœus, U- nanus, Cytisus scoparius, Hedera Helix, Vaccinium Myrtillus, Arbutus Unedo, Andromeda polifolia, Erica Tetralix, E. cinerea, E. ciliaris, E. vagans, E. mediterranea, Calluna vulgaris, Phyllodoce cœrulea, Ligustrum vulgare, Daphne Laureola, D. Mezereum, Hippophœ rhamnoides, Viscum album, Pinus sylvestris, Junipervs communis, Taxus baccata.

– 353 –

I have made a series of experiments to test the amount of transpiration undergone by different plants, and these show interesting results as far as they have gone; but as only a limited number of plants have been examined up to the present, and the whole series seems to be worth concluding, I will not give more than a single example to illustrate the facts.

The method adopted was to take a newly cut leafy twig of Fuchsia excorticata (as a control), weigh it, and hang it up in a dry place for a stated time. Twigs or branches of the other plants to be tested were treated in the same way and for the same period of time. The method is not a very accurate one as a test of transpiration, but it gives comparative results which are suggestive. It only shows the actual loss by evaporation, but gives no correct idea of the amount of water which the plants tested could abstract from a constant water-supply and give out to the atmosphere through their breathing-surfaces. Still, the following facts are interesting:—

In Fuchsia excorticata there are no stomata on the upper surface of the leaves, but as many as 228 per square millimetre on the underside. These are all placed on a nearly level surface, so that when much evaporation takes place the whole leaf tends to wilt and wither.

In Veronica Hectori (a xerophytic form) there are also no stomata on the upper or appressed side of the leaf, but on the lower (outer) side there are about 128 stomata per square millimetre. These are placed in somewhat deep pits, so that when much evaporation takes place the guard cells above them overarch and tend to close the orifice, and the leaves show no sign of wilting.

A small leafy branch of Fuchsia excorticata exposed to dry air for twenty-eight hours lost 66.41 per cent. of its weight; a branch of Veronica Hectori exposed for the same time to the same conditions lost only 8.33 per cent. of its weight. This illustration is typical of the whole series : some lose a greater and some a lesser proportion of their moisture; but in all these 220 species of plants the amount lost by transpiration is very small when compared with a thin-leaved plant like the Fuchsia.

The cause which has led to the development of this desert flora would appear to be the extremely dry conditions which prevail on the eastern side of the main range of the Alps. This is brought about by the moistureladen winds which strike the range from the north-west losing their water as they are forced over the mountain barrier, and descend on the east side as dry, hot winds. The average rainfall along the western side of the Southern Alps probably everywhere exceeds 100 in.; on the eastern side it is in many districts less than 20 in. It is impossible to get exact records, for they have not been kept; but on the Ida Valley Station, in Central Otago, where rain-gauge records were kept for a few years, I am informed the rainfall was always under 15 in. per annum, and in some seasons fell as low as 10 in. This extreme dryness, coupled with the intense winter frosts, has caused the disappearance of nearly all plants except such as were able to withstand these severe conditions. The average height of the main range may be taken as 7,000 ft.: if the range were elevated another 5,000 ft. it would produce a still more severely desiccated region, in which only a xerophytic flora could survive, and it would also account for all the Otago lakes and West Coast Sounds, and all the glacial phenomena required to explain Professor Park's views. Only it was probably much further back than the Pleistocene.