19 September 2008

The oceanic islands of continental New Zealand

It would seem like a no-brainer to suggest that New Zealand is made up of a group of oceanic islands. After all, New Zealand IS a group of islands set in, what can only be described as ocean; with plenty of it in all directions. However, most descriptions about the history of New Zealand emphasize the point that New Zealand is a continental island and therefore very different in origin than oceanic islands. What’s the difference? Oceanic islands are islands that have been generated in mid ocean, usually by volcanic activity. They are emergent from the sea. The Galapagos Islands are a good example of this. Continental islands, on the other hand, were once part of larger landmasses that have become isolated from these continents. New Zealand was geologically once a part of the large continent of Gondwanaland and qualifies as a continental island. Or does it?

A review paper by Julia Goldberg and Steve Trewick (Massey University), with Adrian Paterson (Lincoln University) has challenged this idea. The paper, published in the Philosophical Transactions of the Royal Society B, is part of a special issue on “Evolution on Pacific Islands: Darwin’s Legacy” and is entitled “Evolution of New Zealand’s terrestrial fauna: a review of molecular evidence”. The authors argue that several observations are commonly made about the New Zealand and its natural history: 1) New Zealand is isolated from other landmasses. 2) The biota is unique. 3) There is a high level of endemicity (species only found in NZ). 4) It includes many behaviourally or morphologically strange and distinctive taxa (divaricate plants, strange reptiles, giant birds). 5) The composition of the biota is disharmonic (some groups are over-represented and many are not found in NZ at all, no snakes, few ants and mammals).

These observations have generally been interpreted as the result of a long isolated continental island biota. However, Goldberg et al challenge these points: 1) Physical isolation does not equate to biological isolation (all oceanic islands with no mainland links eventually end up with relatively large biotas). 2) All biotas of distinct areas are, in fact, unique! (That's what makes them distinct!) 3) Endemicity is usually high on islands, including oceanic islands like Hawaii. 4) Distinctive taxa are common products of evolution on islands, like the Dodo and Galapagos finches. 5) Disharmonic biotas are best explained by random colonisation and extinction events (It is often somewhat unpredictable about which species will successfully colonise).

Their conclusion is that New Zealand is no different from most other oceanic islands. Recent geological reappraisals have also suggested that New Zealand (or Zealandia) was inundated about 24 million years ago and then rose out of the sea from about 21 million years ago (see The fall and rise of NZ). If this is the case then, despite the fact that NZ has continental geology, New Zealand is an oceanic island.

Goldberg et al reviewed the molecular data that have accumulated over the last two decades for New Zealand taxa. This data allowed scientists to examine evolutionary relationships of species and the time of divergence of lineages since their last common ancestors. They looked at links between New Zealand and Gondwanan landmasses, Australia, the Pacific, the Chathams, and within New Zealand. The overwhelming pattern of most of these studies is that taxa have arrived in New Zealand since the break up of Gondwanaland (see also The different meanings of Gondwanan) and that diversification is relatively recent. Again this is consistent with New Zealand being an oceanic rather than continental island. This is a major challange to our understanding of New Zealand's natural history and should stimulate debate for years to come.

10 September 2008

It helps to know the local lingo but don't expect the natives to be friendly

Charles Darwin first puzzled over whether native species would help or hinder the invasion of related exotic species. Do natives make it hard for related invading exotic species to get established? Distantly related invaders might have an advantage in not being susceptible to many of the local pests and diseases, and, some may have novel traits that give them a competitive edge over the natives.

On the flip side, Darwin also surmised that species closely related to natives would be most likely to have the right combination of traits to do well in the new environment. When in Rome, don't talk Swedish. Knowing which of Darwin's two scenarios usually occurs in nature could prove useful for predicting new invasive weeds.

Jeff Diez, a postdoc with the Bio-Protection Research Centre at Lincoln University, and colleagues, had a close look at this problem using the invasion of plants into Auckland. Their results were recently published in the journal Ecology Letters.


The Auckland plant invasion is one of the best documented in the world, thanks to the work of several New Zealand botanists, particularly Alan Esler. When Jeff had a close look at Alan's data, he found that Darwin was right on both counts. It depended on the scale at which you examine the problem.

Jeff and colleagues found that plants introduced to Auckland were more likely to go wild ("naturalise") if there were many related plants in the Auckland native flora, but less likely to naturalise if these native plants were abundant.

When an introduced plant had naturalised, its abundance was related to the abundance of related natives. If natives were abundant in the habitat it was invading, the species would itself be less likely to become abundant. However, if related natives were abundant in other habitats in Auckland, the invading species would be more likely to become abundant.

These results are consistent with introduced species that are related to natives being generally better suited to the Auckland environment than more foreign species. It is also consistent with related natives providing "biotic resistance" to this invasion where they are numerous in the habitat being invaded. It helps to know the local lingo but don't expect the natives to be friendly.

Regrettably, this advance is likely to lead to only minor improvements in our prediction of weed invasions. Nearly half of the variation in naturalisation and 90% of the variation in subsequent abundance was found at the species level. This means that at the end of the day, each species is different enough from its relatives to be largely independent in its success as an invasive species. Jeff and colleagues conclude their paper by noting that "ultimately, we suggest that phylogenetic patterns of invasion may provide interesting insights into the organization of ecological communities, but only a limited basis for predicting invasions".