The origin of species: the Australian connection

Drifting proteas or continents? Historical biogeography of the Proteaceae

3 October 2006

Dr Peter Weston
Principal Research Scientist, National Herbarium of New South Wales, Botanic Gardens Trust, Department of Environment and Conservation (NSW), Royal Botanic Gardens

Peter's profile at the Royal Botanic Gardens, Sydney

Proteaceae illustrations from the Australian National Botanic Gardens

Flora of Australia series

Introduction

Dr TJ Higgins: Good evening and welcome to the Australian Academy of Science. This is the second lecture in the Academy's public lecture series for 2006-07. The series, The origin of species, explores the many Australian connections which have contributed to Charles Darwin's ideas. The series will conclude in May next year, coinciding with the 100th anniversary of Carl Linnaeus' birth.

It is a great pleasure to have Dr Peter Weston from the Royal Botanic Gardens in Sydney as our speaker this evening. Peter completed his PhD with Roger Carlin at the University of Sydney and has worked on research projects at Kew Gardens in London and Rhodes University in South Africa.

He has published in a wide range of areas, including an intriguing paper on the conservatism among sexually deceptive orchids. But it is the Proteaceae that are his natural home. It is with great pleasure that I introduce Dr Peter Weston, to talk about Proteaceae and drifting continents. Thank you very much, Peter.

(Click on images for a larger versions)

Dr Peter Weston: Thank you. This slide shows the poster that Rhodes University in South Africa put together for a lecture, which I gave earlier this year. I think the Proteaceae make great floral emblems: the King Protea is the floral emblem of South Africa and Telopea speciosissima is the floral emblem of New South Wales. This evening I will talk about the development of the science; where the Proteacea are and how they got there.

(Click on image for a larger version)

As this series is dedicated to Charles Darwin and his connections with Australia, I wanted to start with a quote from the beginning of the first of his chapters in On the Origin of Species about geographic distribution. This shows that he thought geographic distribution was important for his theory.

In considering the Origin of Species...a naturalist, reflecting on the mutual affinities of organic beings, on their embryological relations, their geographical distribution, geological succession and other such facts, might come to the conclusion that each species had not been independently created but had descended, like varieties, from other species. (C. Darwin, 1859. On the Origin of Species)

He goes on to argue why the facts of geographic distribution, as he puts it, are best explained by the theory of descent with modification, and not from special creation or multiple special creations.

(Click on image for a larger version)

This diagram was published in one of the editions of On the Origin of Species – I'm not sure which one – but it illustrates Darwin's evolutionary model. As well as arguing that natural selection was the most important, or perhaps the sole mechanism by which evolution actually proceeds, Darwin also made a big contribution by suggesting what the shape of evolution was. Basically it involved ancestral descendant lineages and you can see these in the diagram. It shows the extinct species and identifies the hypothetical ancestors that gave rise to the existing species found at the top of the diagram. This was a very important contribution in the mid 19th century, because up to that time there had been many experiments with different ideas about the predominant shape of evolution. As we shall see, the branching points in the evolutionary tree are important with respect to geographic distribution.

(Click on image for a larger version)

The Proteaceae is a family that I have spent many years researching. They are found in a variety of habitats but chiefly in communities on nutrient poor soils that are more often than not well drained. We can see some examples here. Grevillea gillivrayi grows in New Caledonia in scrubland communities similar to those in Australia in which the Proteaceae often dominate. In the northern Andes in Ecuador Oreocallis mucronata – with its creamy white flowers – grows in similar shrubby communities.

(Click on image for a larger version)

The Proteaceae are also found in forest communities in southern and eastern Australia. Many species grow in eucalypt forests: Telopea oreades in Monga State Forest and various species of rainforest Proteaceae in the McDowell Range in north-eastern Queensland.

(Click on image for a larger version)

The Proteaceae grow in these habitats in various locations around the world but they are mostly restricted to the southern hemisphere and, more particularly, mostly restricted to land that once formed part of the super continent Gondwana. This is a very simple map of the distribution of the entire family throughout the world. What it doesn't show are the centres of diversity: in Australia, especially in eastern and south-western Australia; the Cape region of South Africa, especially the Western Cape; and tropical South America and New Caledonia. There are slight extensions into land that is either non-Gondwanic or very ancient fragments which detached from Gondwana hundreds of millions of years ago.

(Click on image for a larger version)

The other 19th century biologist I want to talk about is Joseph Hooker, a friend and close colleague of Charles Darwin's. What I admire most about Hooker is that he was a fantastic lateral thinker and he anticipated certain aspects of Darwin's theory by some years, and he got aspects of evolution processes more accurate than Darwin did.

Many Victorian gentlemen started their biological careers by going on voyages of discovery – Darwin's voyage was on the Beagle between 1831 and 1837. Hooker's voyage, some years later, was around the Antarctic. He served as a botanist, collecting samples and making observations. On his return he prepared the Flora of Australia and Flora of New Zealand and wrote introductory essays explaining various observations of the flora.

He noted in the introductory essay to the Flora of New Zealand that:

There are upwards of 100 genera, subgenera, or other well marked groups of plants, entirely or nearly confined to New Zealand, Australia and extra-tropical South America.

Enough is here given to show that many of the peculiarities of each of these three great areas of land in the southern latitudes are representative ones, affecting a botanical relationship as strong as that which prevails throughout the lands within the Arctic and Northern Temperate zones...

This to me is the amazing conclusion to these observations:

...and which is not to be accounted for by any theory of transport or variation, but which is agreeable to the hypothesis of all being members of a once more extensive flora which has been broken up by geological and climatic causes. (J. Hooker, 1859. Introductory Essay to the Flora of New Zealand)

So in 1853 Joseph Hooker was explaining that the plants spread across different land masses in the southern hemisphere had common ancestors. Not only that, he was also saying that the plants must have been broken up by geological or climatic causes. He produced an impressive list of different plant groups that were distributed in the temperate lands of the southern hemisphere, among which were what we might call the waratah group.

(Click on images for a larger versions)

He drew attention to the fact that Telopea truncata was growing in Tasmania, Knightia excelsa – which looks like a waratah – growing in New Zealand and Embothrium coccineum, the Chilean Firebush, a close relative of these plants, growing in Chile and Argentina. He enumerated scores of taxa which have similar distributions.

(Click on image for a larger version)

Hooker postulated that these distributions could be explained by land connections between these areas and the most likely one in this part of the world would be through Antarctica.

(Click on images for a larger versions)

Another member of the Proteaceae included on his list was the genus Lomatia, represented here by Lomatia ferruginea from Chile and Argentina and Lomatia tinctoria from Tasmania. Once again, Hooker postulated land connections to explain these repeated distributional patterns.

(Click on image for a larger version)

When you put the distributions together you get a coincident pattern across the South Pacific.

However, Charles Darwin disagreed with his friend regarding the explanation for these patterns. Darwin preferred to see all of these distributions as resulting from dispersal across ocean gaps from a centre of origin for each taxon. He said:

If the existence of the same species at distant and isolated points of the Earth's surface, can in many instances by explained on the view of each species having migrated from a single birthplace...then, considering our ignorance with respect to former climatic and geographic changes and various occasional means of transport, the belief that this has been the universal law, seems to me incomparably the safest. (C. Darwin, On the Origin of Species, 1859)

In a sense, you can paraphrase that and say, 'Well, if you don't know anything about the climatic and geographical changes then we know that they must've flown across from one continental mass to another', and:

By these means, as I believe, the southern shores of America, Australia, New Zealand have become slightly tinted by the same peculiar forms of vegetable life.

(Click on image for a larger version)

Darwin acknowledged Hooker's general patterns of distribution in the southern hemisphere but as he saw it, the explanation that best fit these patterns was one of dispersal from the centre of origin. He didn't actually say where, for example, the Waratahs came from but they might have originated in South America and then dispersed across the South Pacific to New Zealand and then to Tasmania.

Over the next 100 years and more there were two competing traditions in biological research. Darwinian biogeographers pursued Darwin's aim of identifying centres of origin and plotting roots of dispersal from those centres across maps of the world. I should emphasise Darwin's dispersal theory meant a world which was geologically fairly stable. He thought that there could have been submerged land bridges and similar geological structures but he wasn't at all interested in the idea of land masses moving around and carrying flora and fauna with them. The dispersal tradition dominated in zoological biogeography or zoogeography right up into the 1970s. On the other hand, biogeographers in the Hookerian tradition looked for general repeated patterns and tried to explain those patterns in terms of geographic and climatic changes.

If we go forward from Hooker more than a 100 years, during which time very little research on the evolution of Proteaceae took place, we come to the late 1950s and early 1960s, when there was a flurry of activitiy from two research groups – one led by Lawrie Johnson and Barbara Briggs who were based at my home institution, the Royal Botanic Gardens in Sydney.

(Click on image for a larger version)

Lawrie, sadly, has since passed away but he and Barbara were very close colleagues for decades. This photograph, taken at a function in the Domain, is the only one that Barbara could furnish that had both of them in it.

At the height of their powers Lawrie was the Director of the Botanic Gardens in Sydney and Barbara was his deputy. In 1963 they published the first monograph on the evolution of the Proteaceae. When it came to explaining the distributions of different groups in the Proteaceae they concluded that:

There is evidence indicating a tropical origin, and therefore suggestions of southern connections between Australia and Africa are discounted, though they may have occurred between Australia and South America. (Johnson and Briggs, 1963. Evolution in the Proteaceae, Australian Journal of Botany, 11: 21-61)

They did discuss the theory of continental drift but geologists had assured them that there was absolutely nothing in it, not a scrap of evidence supported continental drift. So they had to explain the distributions of the Proteaceae by appealing to, among other things, a stable map of the continents.

But then in the late 1960s, deep sea drilling found evidence that geologists could only explain using continental drift. Magnetic anomalies on the ocean floor described stripes across the ocean basins that were parallel to what turned out to be mid-ocean spreading ridges. So by the early 1970s there was a revolution in geological theory and continental drift had transformed from being a crazy idea to geological orthodoxy.

Using several of Chris Scotese's illustrations, which can be downloaded from his website (www.scotese.com), we can go back to the late Jurassic, a suitable starting point for our story.

(Click on images for a larger versions)

We can see in the late Jurassic the super continent Gondwana which accounts for most of the distribution of existing Proteaceae. By the late Cretaceous, Africa, India and Madagascar had departed from the rest of Gondwana, but there was still continuous land between Australia and South America, via Antarctica. New Zealand and New Caledonia formed part of a single continental block still connected by solid land to the rest of Gondwana. By the Cretaceous-Tertiary Boundary, Africa, Madagascar and India had drifted well out into the Indian Ocean. There was still a connection between Australia, Antarctica and South America, and the piece of land comprising New Zealand and New Caledonia had drifted out into the south west Pacific, the Tasman Sea. Note that at this time Antarctica had no icecap. So there was the possibility of continuous vegetation from Australia all the way through to South America via Antarctica and the possibility of overland dispersal through this route.

But then around 40-50 million years ago Australia began to move away from Antarctica and around 30-40 million years ago the last overland connection between Australia and Antarctica, via Tasmania, ruptured and Australia quickly drifted north. Similarly, South America departed from Antarctica and a circumpolar current formed, which effectively isolated the polar region climatically from the rest of the world, leading to the formation of the Antarctic icecap.

Johnson and Briggs updated their monograph of the evolution of Proteaceae in 1975. A lot more was known about the Proteaceae since their first publication and they were able to reconstruct evolutionary relationships within the family with a lot more detail than they had in 1963 and, of course, they now had continental drift to explain distribution patterns. They revised their earlier explanation considerably and concluded:

The Proteaceae...originated before the middle Cretaceous and was dispersed with the land masses themselves in the break up of Gondwanaland. (Johnson and Briggs, 1975. On the Proteaceae – the evolution and classification of a southern family, Botanical Journal of the Linnean Society 70: 83-183)

I have spent most of my research career testing their hypotheses about evolutionary relationships and Proteaceae, testing the Gondwanic hypothesis of Proteaceae by geography. The evidence that we – those of us who have been dealing with the subject – have sought to explain, is the distribution of the taxa. What we find, group after group, is a repetitive distribution pattern of the Proteaceae on different Gondwanic fragments.

(Click on images for a larger versions)

I gave this lecture first in South Africa, so I will deal with the proteas first. Here we have Protea neriifolia, found in South Africa. The closest relative of Protea is another African genus called Faurea, found in Madagascar – separated from Protea by a small ocean gap.

(Click on images for a larger versions)

Another very large group endemic to the Cape region of South Africa, although not particularly closely related to Protea or Florea, is Leucadendreae. It includes genera such as Leucospermum, Mimetes, Diastella and Orothamnus which aren't commonly cultivated here, as well as Vexatorella, Paranomus, Sorocephalus, Spatalla, Leucadendron and Serruria.

(Click on images for a larger versions)

The centre of diversity for all these genera is the Western Cape, although some do stretch into the Eastern Cape. Their closest relatives are the Australian genera, Adenanthos, for example A. detmoldii, and Isopogon. Looking at the distribution map we see the Leucadendreae tribe in southern Africa and two genera distributed across southern Australia.

The relationships that I'm describing are based on a combination of morphological study, the kind of thing that Lawrie and Barbara did, and molecular sequence analysis which provides a large number of similarities and differences with which to assess evolutionary relationships.

(Click on images for a larger versions)

The molecular data has shown that the closest relative of the Australian genus Petrophile, isn't Isopogon but Aulax – a small African genus of four species restricted to the Western Cape. The distribution of the Petrophileae tribe – Petrophile and Aulax – is very similar to that of Leucadendreae.

(Click on images for a larger versions)

Another well known genus of the Proteaceae is the Australian Macadamia. Its closest relatives are found in tropical South America, for example Panopsis, and the Cape region of South Africa, for example Brabejum.

(Click on images for a larger versions)

The sub tribe Embothriinae includes the Australian genera Telopea (waratahs) and Alloxylon (tree waratahs) and the South American genera Oreocallis and Embothrium.

Joseph Hooker used Embothriinae as an example of the repeated southern distribution pattern but he got the taxonomy wrong: he didn't know about Oreocallis and Alloxylon and he included the New Zealand genus, Knightia, which is not closely related to this group. But it didn't make much difference to his story. Looking at the distribution of Embothriinae, there are waratahs in Australia stretching up the eastern coast and going as far north as southern New Guinea, and a couple of genera in tropical and temperate South America.

(Click on images for a larger versions)

The distribution of the next genus, Lomatia, is extraordinarily similar. An example of this genus is Lomatia silaifolia, a species found just north of Canberra all the way up into central east Queensland. All Lomatias have aerodynamic wing seeds. The fruit split open on maturity and the seeds hang out before falling to the ground, rotating like helicopters as they go. The distribution of Lomatia is very similar to that of the close relative the Embothriinae, or waratah group, in eastern Australia. Lomatia doesn't extend into New Guinea but otherwise the distributions are almost identical.

(Click on images for a larger versions)

Orites is another genus that occurs in Australia and temperate South America. O. excelsa is a rainforest species found in the Hunter Valley all the way to North Queensland. There are several species found in Tasmania and one in alpine south-eastern Australia and another species in Chile, O. myrtoidea. Once again the distribution pattern is repeated across the southern continents.

(Click on images for a larger versions)

Hicksbeachia pinnatifolia, the Red Bopple Nut from northern New South Wales and south-eastern Queensland, is closely related to Kermadecia pronyensis from New Caledonia – bringing in parts of Gondwana that Hooker didn't write about. New Caledonia, Vanuatu and Fiji, all have Gondwanic crust contributing to their geology.

(Click on images for a larger versions)

Athertonia, the Atherton Nut, is another genus from north-eastern Queensland and its closest relatives are the genera Virotia from New Caledonia and Heliciopsis from South-East Asia, the only genus that occurs wholly on the Asian side of Wallace's Line. This area of South-East Asia is very ancient Gondwanic – a fragment that departed from the north-western coast of Australia over 250 million years ago.

(Click on images for a larger versions)

The genera Catalepidia from north-east Queensland and Malagasia from Madagascar show another connection.

(Click on images for a larger versions)

The tribe Persoonieae is represented in Australia by Persoonia, a large genus with about a hundred species. Persoonia has relatives like Garnieria spathulifolia in New Caledonia and Toronia tora in New Zealand.

(Click on images for a larger versions)

The genus Stenocarpus, which includes the Firewheel Tree, S. sinuatis, is related to other genera that are endemic to Australia and many that occur in New Caledonia. In fact, the genus has its greatest diversity in New Caledonia, however none as spectacular as S. sinuatis. So, once again we have this connection between two Gondwanic fragments, New Caledonia, and Australia.

(Click on images for a larger versions)

Even the Grevillea genus – which is very diverse in Australia – occurs elsewhere, including three species in New Caledonia, for example G. gillivrayi. A closely related genus, Finschia, goes right out into Micronesia and Palau Island in Vanuatu. The actual species that occurs in this part of the south-west Pacific is a nut crop for Melanesians and they probably moved that species around in recent times.

There are an impressive number of repeated distribution patterns like those that Joseph Hooker used as evidence to propose the idea that distribution patterns result from geographic and climatological changes.

But a scientist's work is never done. We don't just rest on our laurels and say, 'Okay, we've sorted the distribution and evolution of Proteaceae'. We're continually trying to test our hypotheses more vigorously than we have in the past because we can never say that we actually have the truth. The work that I've done on the Proteaceae has mostly been in collaboration with various colleagues, one of whom is Professor Michael Crisp, a researcher here in Canberra at the Australian National University.

(Click on image for a larger version)

We studied the phylogeny of the Waratah group and its close relative, Lomatia, back in the 1980s and nineties and we produced a phylogenetic tree.

This diagram is similar to the diagram from On the Origin of Species by Darwin. But Darwin's diagram was hypothetical. He said, 'Well, this is what I think the shape of evolutionary processes is like'. We constructed an evolutionary tree for Emobthriinae using morphological similarities and differences. The numbered spots, crosses and equal signs indicate postulated changes in morphological feature. We can infer that a particular ancestor would have a particular set of characteristics and all of the descendants of that ancestor evolved from that and accumulated various changes. That is our hypothesis to explain the morphological variation in this group.

A powerful feature of this diagram is that you can test it by looking at another set of features, construct the same kind of tree and see whether it gives the same answer. It's powerful because there are many different ways of rearranging the branches, and getting the same answer by chance is highly improbable. We tested this tree using DNA sequence data and with a couple of exceptions we got exactly the same tree. It tells us that we're on the right track in constructing the evolutionary relationships of these plants.

We used this diagram to develop an hypothesis about biogeography. In the Hookerian route and the Darwinian tradition, we replaced the species labels with the areas in which the species occur, to develop a hypothesis about the way ancestral distribution has been split up by geological and climatic causes. We also analysed the relationships amongst the species of Lomatia, a genus closely related to Embothriinae with an almost identical distribution. Similarly, we replaced the species labels of the Lomatia with the locations where the species occur. The resulting diagram is the best estimate of area relationships that we can tease out of the phylogenic trees for both of the groups. Interestingly, it makes a lot of sense geologically and geographically: South America splits off from the Australasian areas first, about 40 million years ago, and some time later Tasmania splits from the mainland for the first time with the sea flowing into Bass Strait for the first time. This is followed by New Guinea splitting from northern Australia and so on.

(Click on image for a larger version)

The fossil record has also been used to test the Gondwanic hypothesis of Proteaceae. This here is a fossil pollen grain. Most fossils from plants are actually spores and pollen grains because they have extremely resilient cell walls that fossilise well. The oldest pollen grain attributed to the Proteaceae family is Triorites africaensis – 94 million years old, found in Brazil and Western Africa.

(Click on images for a larger versions)

This 70 million year old pollen grain had been attributed to the genus Telopea. We included pollen features in the analysis of the Waratah relationships and we concluded that this fossil isn't from Telopea but it is a member of the Embothriinae sub tribe. This particular fossilised pollen grain is indistinguishable from what we reconstruct as the ancestor of the whole sub tribe Embothriinae.

(Click on images for a larger versions)

When we compare it to Telopea truncata or Alloxylon pinnatum we see that they are very similar. We do use fossil evidence for looking at distributions in the Proteaceae, more critically than we have in the past, but we have to be cautious about our interpretations.

(Click on image for a larger version)

Some fossils are very impressive. This inflorescence of a fossil found in south-eastern Australia is almost indistinguishable from the inflorescence we would interpret – from our analysis – as being an ancestor of the sub tribe Musgraviinae, the closest relative of the Banksia group. The Musgraviinae are rainforest trees found only in north-eastern Queensland but this fossil came from Anglesea in Victoria. This shows that distributions within Australia have changed dramatically over time. Some researchers use fossil evidence to try to locate centres of origin and draw dispersal maps out from those centres and across the southern continents. I would argue that this is not a good way to reconstruct evolutionary history because it relies on a number of assumptions, some of which we know to be wrong, such as the completeness of the fossil record. But fossil evidence is nevertheless crucial if we want to further test the Gondwanic theory of proteaceous biogeography.

(Click on image for a larger version)

I'm going to show you results from some of the work I've been doing with Nigel Barker from Rhodes University, my host in South Africa earlier in the year.

(Click on image for a larger version)

This phylogenic tree was constructed using molecular dating – aligning and analysing DNA sequences – of different species of Proteaceae from many different genera. The methods we use for doing this kind of analysis are similar to those for analysing morphological similarities and differences. If you will recall, with the morphological tree for the Waratahs some branches had more changes plotted on them than others. In the molecular tree this is represented by the branch lengths.

Molecular dating is based on the theory that we can model the differences in branch length as a result of two different processes. One is the rate of change – random changes – in the molecules that are used to create the tree. On this basis we find that there are different rates of change along different branches. The second process is the evolution of the rate of change of cells. Methods have been developed that attempt to optimise these two sources of differences in branch lengths. The aim is to find points that are comparable along the different branches under an assumption that, rates in molecular change can give us something approximating a clock. So similarities between these different species, according to the notion of the molecular clock, decay in relation to time. The rates at which the similarities decay may vary from branch to branch but we can still use these methods to try to estimate comparable time points along the different branches.

(Click on image for a larger version)

The result of this analysis is a tree, with a timescale along the horizontal axis and all the species lined up along the right-hand side. We calibrated the tree – to give an absolute timescale – using a number of fossils. For example, this particular calibration point [F] represents the 70 million year old fossilised pollen grain that I showed earlier. Some branches were stretched and others were compressed to enable the branches to line up with the correct time on the horizontal axis.

We expected this analysis to corroborate most of the elements of the Gondwanic period of proteaceous biogeography. However, there is only one member of Leucadendron (from Africa) and its closest Australian relative, Adenanthos in this analysis and it suggests they diverged about 40 million years ago. So, an African and an Australian genus in the Proteaceae shared a most recent common ancestor as recently as 40 million years ago. Another group, Macadamia (from Australia), and its African relative, Brabejum, share a most recent common ancestor about 50 million years ago. Panopsis, the tropical South American genus related to Macadamia, shares a most recent common ancestor with Brabejum 40-50 million years ago.

(Click on image for a larger version)

The last overland connection between Africa and any other part of Gondwana was with tropical South America about 105 million years ago and as recently as 66 million years ago Africa was out in the middle of the Indian Ocean. Clearly, our estimate of the age of separation between Leucadendron in the Cape region of South Africa and Adenanthos in Australia at 40 million years ago is too recent to be accounted for by continental drift. Similarly with Macadamia in Australia, Brabejum in Africa and Panopsis in South America the last connection between Australia and South America was about 40 million years ago. This fits in with the estimate of the age of the Macadamiianae but it doesn't explain the occurrence of Brabejum in the Cape region of South Africa.

(Click on image for a larger version)

Our conclusion is that either the dating method used is fundamentally flawed – the assumptions are completely wrong and we shouldn't be doing it – or the Proteaceae has dispersed over water. This is the first time in 30 years that this has been suggested and we are now investigating processes that could have assisted this dispersal, such as west wind drift and polar currents.

In terms of dispersal, a concept discussed in great detail by Darwin, some groups in the Proteaceae really can be explained as having flown across the Indian Ocean. For example, Petrophile fruits are small, light and very aerodynamic. If you pick them out of the fruiting cones they drift away on the breeze. Similarly, the seeds of Leucadendron and most of its relatives are also aerodynamic. So it is possible to imagine the ancestors of Aulax getting to Africa from Australia or the ancestors of the Leucadendreae getting to South Africa from Australia.

(Click on image for a larger version)

If you look at the fruit of the Macadamiinae, we have macadamia nuts, with which you'll all be familiar, but the fruits of other members of the Macadamiinae sub tribe are just like macadamias. They are ballistically dispersed, fall like stones from the tree and have a very short viability. The fruit of the Brabejum, the wild almond from the Cape region, will only last a couple of weeks if it is allowed to dry out. Horticulturally they're quite challenging to get into Australia before they lose their viability. Can we sensibly speculate that the ancestors of this plant got to Africa either by flying across the Atlantic from a shared ancestor with Panopsis or across the developing Indian Ocean from a shared ancestor with Macadamia?

Now, one of the reasons for instigating this series of lectures is to counter the idea of intelligent design. Some people might say, 'You can't explain the distribution of the Macadamiinae. Your best analysis suggests that this group is far too young to have got to where it lives now by riding around on drifting continents. The fruits just can't be explained in terms of long distance dispersal across large ocean gaps. Shouldn't you just invoke a supernatural explanation?' Of course, the answer is that a supernatural explanation is an ultimate answer and science isn't about ultimate answers. If we just say, 'Okay, well, God did it,' then the problem is solved. We need not continue any further hypothesising or testing ideas about the evolution of this group or its biogeography. It is unscientific. If we had done that since the year dot we would still be living in caves.

That's my little sermon about intelligent design and its inapplicability or incompatibility with science. I would like to conclude by saying that Hooker was right about his Proteaceae, about the Waratahs, although he wasn't quite right about their taxonomy, and Darwin was probably right about some of the others. Thank you very much.

Dr TJ Higgins: I'd like to ask you all to join me in thanking Peter for a most intriguing and interesting talk. Thank you very much, Peter.