Metrosideros angustifolia (Myrtaceae): the Pacific face of fynbos

Mainland floras, such as that of the ancient and stable continent of Africa, combine many lineages of plants in a plethora of orders, families and genera.

Hence we have the floristic richness of the southwestern Cape (, where not only are fynbos ( and succulent karoo ( extremely diverse within themselves, but several distinct floras (fynbos, Karoo, forest, thicket, and even savanna) can share a single mountain range.

Oceanic Islands, by contrast, have relatively few lineages. This is because only a few lucky plants manage to cross the vast barriers of the sea, drawing on their genetic plasticity to fill the vacant niches offered by isolated mountains erupted from the ocean floor.

Or, at least, so we learn from textbooks.

Metrosideros ( and is a classic example of an island genus, having radiated widely in the Pacific.

Yet Metrosideros angustifolia ( pops up in what seems to be the most unlikely of places. It is an integral part of the Fynbos Biome - as far as possible from the Pacific, and where every niche of shrubs seems already to be filled by diverse orders, families and genera.

Metrosideros angustifolia, restricted entirely to the western part of South Africa, is a nondescript shrub. It gives an impression of anonymity among the many narrow-leafed, semi-sclerophyllous evergreens of the Fynbos Biome.

The distribution of this species is unnacountable for a genus otherwise restricted to Pacific Islands; it occurs along the streams of the southwestern Cape, with an outlying population on the Orange River.

We have no idea how Metrosideros reached Africa, or why it succeeded among the hundreds of shrub genera already crowding Western Cape province.

Apart from this species, all of the 50-odd spp. of Metrosideros originate from New Zealand (, New Caledonia (, New Guinea (, and the Philippines ( in the west, to the Marquesas ( and Hawaii ( in the east.

Even Australia lacks any indigenous member of this oceanic genus. This is despite the facts that this island continent borders the Pacific, and is dominated by various Myrtaceae.

Most spp. of Metrosideros are at home on volcanic soils. By contrast, M. angustifolia is unknown from the volcanic parts of South Africa, preferring wet, acidic soils in the form of the stream-gravel of sandstone ( of Western Cape province.

Gardeners in South Africa know Metrosideros not as the indigenous shrub, but as an eye-catching tree, Metrosideros excelsa ( and

The red blooms attract sunbirds ( in South African gardens. But how many readers appreciate that it is not only birds, but also geckos (Hoplodactylus, that pollinate M. excelsa on its islands of origin in New Zealand?

There, it can grow so close to the shore that its roots are bathed in seawater ( and

The species most renowned among botanists is Metrosideros polymorpha (, which dominates the natural forests of some of the most remote of islands. This Hawaiian species shows the remarkable plasticity of form in Metrosideros, shape-shifting to fill otherwise empty niches.

Metrosideros polymorpha is, indeed, one of the most variable species of plant on Earth.

It can be

with accordingly variable leaves over the full range of altitudes on Hawaii and its associated volcanic islands.

The following show the remarkable range of variation in the appearance of the leaves of M. polymorpha:
( and and and and and and and and and and and and and and and and and and and and and and and and and

The following show the aerial roots of M. polymorpha:
( and and and and and and and and and and

Metrosideros polymorpha can survive being buried in lava, regenerating its roots once the volcanic ash has cooled ( and

Other spp. of Metrosideros are creepers or lianes.

Metrosideros perforata ( clings tightly to bark by means of sucker-roots as it climbs trees in the form of a delicate ivy. Then the same species lets it all hang down when mature, as an unrecognisably different woody liane of thick 'monkey ropes'.

Metrosideros robusta of New Zealand is a strangler, an island reinvention of the strangler figs (Ficus, of tropical mainlands.

It germinates as an epiphyte in the canopies of trees such as Podocarpus (, the Pacific relatives of the tallest trees in South Africa. It then sends down aerial roots that eventually reach the ground, coalescing on the way to form a hollow bole that gradually extirpates and replaces the host tree.

Once it becomes a free-standing tree, M. robusta itself becomes festooned with various non-strangling epiphytes.

Such 'hanging gardens' are typical of New Zealand, which had few arboreal herbivores before the introduction of Trichosurus vulpecula ( from Australia.

Metrosideros has not penetrated the most similar habitat in Africa: the island-like patches of forest of Podocarpus ( found from Knysna ( northwards to Ethiopia (

I can only speculate that this exclusion has been partly owing to the pressures exerted by Papio (, Dendrohyrax (, and Loxodonta (

Stranglers are not parasitic in the sense of mistletoes. They grow from the canopy down, not from the ground up, and initially absorb nutrients from dust by means of aerial roots. They produce wood as these roots grow down the bole of the host, consolidating and merging as the plant finally finds a rooting medium in the ground.

It may be unremarkable that figs fashion a bole of their own in this way, because their wood is so soft ( that the distinction between root and stem is academic.

However, nobody understands how Metrosideros manages to form hard, dense heartwood within the initially cylindrical bole formed by roots encompassing the host tree.

Metrosideros excelsa is not an epiphyte or a strangler, but produces aerial roots when it becomes a mature tree (please see comment below).

This and other species of Metrosideros often sport the genetic mutations of variegated leaves ( and Further evidence of genetic plasticity is that variegated trees in gardens often revert to the normal green colour on individual branches - as if in control of the mutation.

Metrosideros umbellata ( and, although rarely a strangler, is also weird.

In the Auckland Islands (, far to the south of the main islands of New Zealand, forest of this species presents a 'chaotic spectacle of gnarled, leaning trunks...Massive branches twist, dip and wind for metres. Trunks may rise a short distance from the surface, bend at a right angle to extend parallel to the ground for 5-10 m and then grow vertically to reach the canopy. Serpentine trunks loop along the ground' - in the words of scientists Matt McGlone ( and John Hunt.

The following show this to some extent: and and and and

The South African species is odd in that

  • it has reverted to 'normality', and
  • it has abandoned the showy inflorescences that are typical of the genus.

Metrosideros angustifolia gives no hint of its genetic legacy as a creeper, liane, epiphyte, strangler, or bearer of aerial roots. Not in evidence is the dwarfism seen in the bonsai-like shrubs, or the gigantism seen in the tall or stout trees of Hawaii and New Zealand.

Even the pollination by vertebrates, typical of Metrosideros ( and many other members of Myrtaceae, has been lost. This is particularly surprising in a land of proteas ( and sugarbirds (Promerops,

Instead, M. angustifolia has dull flowers pollinated by insects ( and

Conforming to the unremarkable shrubs such as Morella ( and, Brabejum (, Brachylaena (, Searsia, and, and Freylinia (, that share its streamside scrub among the Cape mountains, the genus has used its remarkable genetic plasticity to abandon weird growths and pretend to extreme ordinariness.

Many plant spp. introduced by humans from other lands are now invading South Africa as if they belong here ( Botanists see these as triffids, and respond with the chainsaw.

However, what if such colonisations have always been business-as-usual in the fullness of ecological time?

Although Metrosideros is capable of being a tall tree, it has descended to the common denominator of the fynbos shrub. And although it is capable of being an aberrant shrub, it has settled for being so mundane that its extreme relatives have, to this day, remained unrealised by many naturalists interested in fynbos.

הועלה ב-יוני 28, 2023 04:44 אחה"צ על ידי milewski milewski


פורסם על-ידי milewski לפני 12 חודשים
פורסם על-ידי milewski לפני 12 חודשים
פורסם על-ידי milewski לפני 12 חודשים

I find it remarkable how these taxon are able to be so biogeographically (e.g., distributions spanning from Australasia to the Cape of South Africa) and morphologically variable (e.g., organisms belonging to the same species growing to visibly different tree and shrub forms). It's still strange (in my perspective) to see willows (Salix spp.) dwared as insigificant shrubs on Ellesmere Island while growing as prominent and large trees near freshwater lakes in local urban parks.

פורסם על-ידי paradoxornithidae לפני 12 חודשים


Two patterns are involved.

The first is plasticity in plant size (which is seen mainly in the height of plants), and the second is plasticity in growth-form (e.g. liane vs free-standing).

The first pattern is remarkably common in plants (unlike animals), and can be seen in most floras. The typical example is the species or genus of tree that grows here as a mere shrub 3 m high, there as a tree 10 m high, and somewhere else as a tall tree, 40 m high: Podocarpus (, Eucalyptus (including particular species of this genus, e.g. E. marginata, see comment below), and thousands of others - including Metrosideros, and indeed including M. angustifolia.

The second pattern is uncommon, and is exemplified particularly by Metrosideros.

The first pattern is a mainstream one, in the floras of mainlands. The second pattern occurs most noticeably on remote islands.

פורסם על-ידי milewski לפני 12 חודשים
פורסם על-ידי milewski לפני 12 חודשים
פורסם על-ידי milewski לפני 12 חודשים
Metrosideros excelsa is not an epiphyte or a strangler but, for unknown reasons, produces aerial roots when it becomes a mature tree.


פורסם על-ידי tonyrebelo לפני 12 חודשים


Here is a good example of the first pattern, which I mentioned above.

Within the genus Cupressus, there is variation in the height of mature individuals, from 20 cm in Cupressus goveniana ( to 102.3 m in Cupressus gigantea.

The environments involved are:

In the case of C. goveniana, extremely nutrient-poor soils in coastal California (see and ; this is a case of 'edaphic stunting' under ample rainfall.

In the case of C. gigantea, the extremely deep canyon of the Yarlung Tsangpo river in southeastern Tibet (; this is a case of extreme protection from fire and other disturbance, in an extraordinary landform.

Please note that, in the case of C. gigantea, the same species can grow to 50 m high in favourable situations.

So, Cupressus exemplifies plasticity in plant size (more particularly height) both within the genus, and within a given species.

The variation is so astonishing that it bears repeating for emphasis. Mature (even possibly senescent) individuals can be as short as the length of your hand, or as tall as a 30-storey building - equivalent in height to the renowned redwood (Sequoiadendron giganteum,

At the lower extreme, the plant can hardly be described as a shrub, let alone a tree. At the upper extreme, the tree is so tall that one cannot directly appreciate its height by standing under it and looking up; it defies the human sensory system. And all within a single species/genus.

פורסם על-ידי milewski לפני 12 חודשים



Many thanks for presenting a photo in the comments, in the way you did. More such would be welcome in my various Posts, on various topics.

The following is perhaps the most readable and informative account that I have seen, of Metrosideros excelsa:

In particular, an explanation is given, on pages 2-3, of how aerial roots help the species to colonise bare ground, in what sounds like a 'triffid-like' way.

On page 5, we learn that M. excelsa is odd, even within its genus, in combining dense wood and unexpected longevity with a pioneering role, in which collapsing, adventitiously rooting branches result in a marching propagation of each individual plant across the ground surface, by vegetative means.

(The word 'marching' is mine, not the author's; I can imagine the 'locomotion' of the plant in time-lapse footage.)

פורסם על-ידי milewski לפני 12 חודשים

I'm only familiar with M. polymorpha, and it is a fantastic tree here in Hawaii. it's also threatened two pathogenic fungi (ROD) which have been killing trees by the tens of thousands on Hawaii island, where M. polymorpha is usually the dominant tree in native forest.

Very interesting read about M. robusta

פורסם על-ידי kevinfaccenda לפני 12 חודשים

Metrosideros umbellata ... "Near Auckland, forest of this species ... " I think you may mean Auckland Islands, which is in the subantarctics, not near Auckland region in the North Island?

Another useful note to make is that most observations of M. polymorpha cultivated outside of Hawaii, are, in fact M. collina - M. polymorpha has been over reported and M. collina has been under-reported in cultivation due to misidentifications.

פורסם על-ידי murray_dawson לפני 12 חודשים


Many thanks for pointing out my inexcusably ignorant error. I have corrected the Post accordingly.

In your experience with the indigenous woody flora of New Zealand, which genera best illustrate the plasticity in plant size (particularly height) that I have described (using the phrase 'the first pattern') in my comments above?

פורסם על-ידי milewski לפני 12 חודשים



Eucalyptus marginata ( is a particularly noteworthy example of 'the first pattern' of plasticity, that I mentioned above.

This is because a) it varies from a multi-stemmed shrub only 3 m high to a tree 50 m high, and b) it occurs in an exceptionally rich area floristically, namely southwestern Western Australia ( and and and, where there are so many species of trees and shrubs, including within the genus Eucalyptus itself ( and,species%20and%20subspecies%20in%20total).), that there seems little reason for one species to shapeshift into different niches in this way.

Eucalyptus marginata, multi-stemmed shrub, 3 m high:

Eucalyptus marginata, tree, 40 m high:

Eucalyptus marginata, tree, 50 m high (please note the human figures for scale, at the base of the tree, in the black-and-white photo):

פורסם על-ידי milewski לפני 12 חודשים
פורסם על-ידי milewski לפני 11 חודשים

@milewski Many thanks for this information

פורסם על-ידי paradoxornithidae לפני 11 חודשים

@milewski "plasticity of plant size" in NZ genera... Leptospermum and Kunzea (also of the Myrtaceae) show great range of plant sizes from prostrate or dwarf plants to tall trees.

פורסם על-ידי murray_dawson לפני 11 חודשים


Here is another example of the first pattern of plasticity.

The family Phytolaccaceae contains only a few genera, most of them obscure.

One genus, viz. Phytolacca, is familiar.

Phytolacca contains 25 spp., naturally distributed mainly in central America but also the Oriental region, the Pacific, Africa and Asia Minor, and North and South America.

All spp. are herbaceous (perennials about 1 m high, with a tendency to stem-tubers) except for Phytolacca dioica, which can grow relatively enormous.

Phytolacca dioica, when mature, is about 15 m wide and high, with a stem 15 m wide just above ground level ( It achieves this size without possessing wood.

Please see

פורסם על-ידי milewski לפני 10 חודשים

I quote the following, on the topic of Metrosideros, from pages 114-118 of Carlquist S (1970) Hawaii: A Natural History ( and

Metrosideros "evolved into a series of types suited to particular localities. A shrubby one...with quaking leaves developed on wind-swept ridges. Types with greatly increased height have developed to suit wet forest; scrubbier forms have been produced in response to conditions in relatively dry ridges. Kinds... with round hairy leaves...tend to occupy lava flows. The hairiness of leaves may counteract the bright, dry conditions more effectively. A low tree with cup-like leaves, rusty-haired underneath, occupies exposed wet ridges...In bogs, Metrosideros is exceptionally small; flowering individuals can be as little as six inches tall...In this way, the Hawaiian (Metrosideros) form a continual occupation of every sort of ecological opportunity from moderately dry forest to bog...(Metrosideros) is the dominant tree of the Hawaiian forests, and it may hold this position precisely because it has such extraordinary variety"

On the same pages, illustrations have the following captions:

"There is no such thing as a typical...Metrosideros polymorpha...The one shown here...a moderate-sized, rounded tree, with smooth elliptical leaves and clusters of red flowers...found in a fairly wet rain forest."

"The 'trembling' type of leaf is found in certain populations...from windy ridgetops on Oahu...These plants have leaves borne in slender petioles".

"...a curious (Metrosideros polymorpha)...grows on high cloudy ridges of the Koolau Mountains, Oahu. It has small cup-shaped leaves covered beneath with dense reddish wool".

"Metrosideros polymorpha can be a very tall tree. Some individuals, as this one, for example, have yellow flowers rather than red ones".

Metrosideros polymorpha "can be diminutive shrubs, consisting only of a few short twigs, when they grow in bogs".

"White, hairy young leaves characterise some trees of (Metrosideros polymorpha) especially those in drier localities (summit ridge, Waianae Mountains)".

Caption to colour photos:
"Metrosideros...dominates the wet phenomenal in its variability: in some places, no two trees are alike".

פורסם על-ידי milewski לפני 10 חודשים

הוספת תגובה

כניסה או הרשמה להוספת הערות