The area of the Hawaiian islands is much less than that of New Zealand but their isolation is much greater, since the nearest continent, North America, is 4000 km distant and both Australasia and Asia are about 7000 km distant. The Hawaiian islands are entirely volcanic and have arisen sequentially on a north-west to south-east line over the last 4-5 million years. Hawaiʻi, the southernmost island, is thus the youngest and is also the largest and highest (4000 m). There seems little possibility, geologically, that these islands have ever been connected to a continent, so botanists and zoologists alike envisage their being stocked by chance immigrants with subsequent evolution of new species. Thus it has been estimated that 272 immigrant species could have evolved into the present approximately 1200 native higher plant species of Hawaiʻi. The distinguishing characteristics of floras that have been derived in this way can be outlined as follows:

Getting There

Not unexpectedly, plant and animal groups with good dispersal ability are strongly represented, while those with poor dispersal ability are absent. Among plants, those with seeds that have special modifications for dispersal (for example: flotation devices for sea transport; plumes of hairs for wind dispersal; hooks for attachment to birds' feathers; or hard seeds in berry fruits eaten by birds and eventually excreted intact), have a good chance of eventually reaching an isolated island. Plant groups which are notable for not reaching such islands, because their seeds are large and/or unspecialised, include conifers; the wind pollinated trees (including oaks and beeches, so important in some temperate forests of the northern hemisphere); and most of the primitive woody flowering plant families in the order Ranales, of which Magnolia is the best known example. Prominent among animals on isolated islands are those with wings — birds and insects — while mammals, amphibians and land reptiles are absent.

Strangely, many species of small, isolated islands have lost or have only vestiges of the dispersal mechanisms possessed by their continental relatives. Thus there are many flightless island birds and insects. Among plants, species belonging to groups that are normally wind dispersed may have seeds with quite inadequate vestigial wings or plumes of hairs. In explanation of this it has been suggested that, although a good dispersal mechanism is necessary for a plant species to reach an isolated island, once the plant is established, good dispersability would become a disadvantage for most of the seeds would end up uselessly in the sea!

Adaptive Radiation

Many genera on isolated islands exhibit a much wider range of growth forms and occupy a much wider range of habitats than the same or related genera on continents. Thus on mountainous islands like Hawaiʻi, some species of a genus may be small and herbaceous and occupy coastal and lowland open habitats, other species may be small trees in wet forests at various altitudes, and yet others shrubs or herbs in alpine vegetation. Carlquist,^{3, 4} who has closely studied island floras and faunas, suggests that the first species of a plant genus to arrive is likely to be a weedy herb, because in general weeds are good dispersers and hardy enough to tolerate the raw open conditions of a new volcanic island. In such a situation many habitats will remain unoccupied for some time (particularly, Carlquist suggests, those suitable for moist forests), so any variants of the weedy coloniser with habitat requirements different from the parent population would have a very good chance of establishing in an unoccupied niche. In turn a variant population could give rise to other variants to occupy further niches including, if the island is high enough, those of alpine habitats. Such a burst of speciation into unoccupied habitats has been termed 'adaptive radiation'. It is not suggested, however, that a species increases its ability to give rise to new varieties and species on migrating from a continent to an isolated island. On the continent it would also form variants, but these would rarely find unoccupied habitats suited to their needs.

Carlquist's view that weedy herbs are frequently the first colonisers of isolated islands implies a general trend in plant evolution on some islands from herbs to trees. This is entirely possible, but it would be an unusual direction for evolution to take, if the generally accepted idea that herbs are advanced, relatively recent in origin and derived from page 16woody ancestors, is correct. In opposition to Carlquist's view some botanists believe that tree species on isolated islands, belonging to genera or even families which are otherwise mostly herbaceous, are primitive ancestral forms that have survived there because of the mild oceanic climates and reduced competition.⁷ On continents, particularly those of the northern hemisphere, woody species of these groups have generally given way to herbaceous species better suited to the strongly seasonal climates which have developed at higher latitudes on continents in recent geological time.

Sexual Patterns

It would seem obvious that a plant species with self-fertile hermaphrodite flowers would stand a better chance of establishing on an isolated island than an hermaphrodite species which is self-sterile or a species with separate male and female plants, a condition known as dioecism. In both the latter cases one plant would not be sufficient to form a population; at least two would be required — one male and one female in the case of a dioecious species — and they would have to establish at the same place and coincide or overlap in time. In some cases, however, this might not be unlikely, as in some species seeds tend to be transported in groups. Berries often contain several small seeds, so a bird might eat a number of berries of a particular species, transport them internally and deposit them in a group on an isolated island. This could be the case with the dioecious, berry-fruited genus Coprosma, so strongly represented in both the New Zealand and Hawaiian floras. Sticky or barbed seeds which attach themselves externally to birds or seeds in mud might also be transported in groups, but this would be less likely with seeds conveyed by winds or ocean currents. Nevertheless, the odds would seem to favour self-fertilising plants as colonisers of isolated islands, and we would expect a lower percentage of say dioecious species on such islands than on a continent. Paradoxically, the reverse is the case. In Hawaiʻi it is estimated that 27.5 per cent of the species are dioecious,³ a much higher proportion than for the United Kingdom, for example, where it is estimated only 2 per cent of species are in this category.

How can this be? If dioecious species are less likely to colonise isolated islands and yet are found there in relatively large numbers, there must be some circumstance which especially favours both the page 17survival and diversification of the relatively few dioecious colonisers and the evolution of dioecious from hermaphrodite island species. This implies that dioecism confers some special advantage in the isolated island situation. It has been suggested that this advantage results from the obligate outcrossing of dioecious species, which brings about an increase in variability. Variable species are more likely to be successful on isolated islands as they are in a better position to take advantage of the unoccupied habitats available, and more likely to survive drastic environmental changes such as those resulting from the succession of glacials and interglacials in recent geological time. On a continent stretching from high to low latitudes, species can survive such fluctuations by migrating with their preferred climates as these climates move towards and away from the equator. On an isolated island, this option is greatly restricted, so with a drastic environmental change, a greater proportion of relatively invariable inbreeding species will become extinct than of more variable outcrossing species, which are better able to adapt to change.

Hybrids

A similar explanation has been suggested for the high level of natural hybridism on isolated islands.⁸'⁹ The progeny from such hybrids displays an even greater variability than that resulting from dioecism, and should thus be an even richer source for the populating of unoccupied habitats or new habitats resulting from environmental change.

Lack of Brightly Coloured Flowers

On islands like Hawaiʻi an unusually high proportion of the native plants have small, shallow flowers which lack bright colours and are pollinated by wind or unspecialised short tongued insects. The exceptions are certain bird-pollinated species which have larger red to yellow flowers of tubular form.

The general lack of brightly coloured flowers is attributable to the absence from isolated islands of the more specialised insect pollinators, particularly long tongued bees, which are attracted by bright colours and pleasant perfumes.