Our trip to the atoll was organised by John Jackson, an American friend who was later to help with the publication of Corals of the World. We knew that Jacques-Yves Cousteau had visited the atoll in 1978 to make a film about its incredible human history, but he found so many ultra-aggressive sharks there that he, of all people, had to abandon diving. So on our first dive we entered the water very carefully. No sharks. In fact we didn’t see a single shark the whole trip; Mexican shark-finners had done a depressingly thorough job.
As expected, there were very few species of coral on the atoll, but the few that were there made a thriving if singularly weird array of coral communities. I collected seven species of coral during my first dive, and that turned out to be the total inventory of all I saw the whole trip. I’d read several theories centred on the notion that reefs had to have a high species diversity in order to flourish or even exist. Not so: this atoll is built almost entirely of Porites, as were the very last reefs of the Tethys.
With not much coral work to be done I had time to have a good look over the atoll itself, which is shaped like an enormous doughnut. The interior lagoon is brackish and completely surrounded by the island proper, which has just one prominent feature, a 30-metre-high chunk of volcanic rock complete with a couple of caverns.
The place had an incredible history. In 1914 a Mexican garrison of about a hundred men, women and children were sent to the island to secure it as Mexican territory. The island had one other occupant at that time, a large, delusional African-American lighthouse keeper by the name of Victoriano Álvarez, who kept the rock for himself. After the outbreak of World War I the Mexicans forgot about the garrison and by 1917 all the Mexican men had died of starvation and disease. But amazingly, some fifteen women and children were still alive, all suffering from scurvy and the children severely deformed from rickets. Once they had lost the protection of their men, the women were enslaved by Álvarez, who took them, one by one, to his rock, raping and beating some and shooting others.
In a story with few rivals, the widow of the garrison commander and her maid managed to club Álvarez to death on the very day a rescue party, in the form of an American warship, arrived. The women left Álvarez’s body to the orange crabs that abound on the island, and the captain of the warship kept his mouth shut lest the women be accused of murder. One of the children who survived subsequently returned to the island with Cousteau’s expedition, as the star of his film Clipperton: The Island Time Forgot.
Much later I talked to Cousteau about this trip and much else besides, especially the discoveries he’d made and his endearingly heartfelt love of the sea. I never thought much of his movies, though they were groundbreaking for the time, but I sure took to the person, a deep-thinking and gracious old man. I would love to have known him better, it would have been such a privilege.
Coral community at Clipperton Atoll, built almost entirely of Porites.
Victoriano Álvarez’s home inside the rock on Clipperton Atoll. The bright orange crabs that finally ate him fill crevices in the rock and are scattered over the sandy floor.
A different evolution
As I’ve previously noted, once I had a good grip on the taxonomy of the corals of the Great Barrier Reef, the more I studied those in regions elsewhere, the more I questioned the details of my original work. The problem wasn’t just to do with geographic variation within species, it was that species appeared to converge in some places and diverge in others. I found this hard to explain to myself, let alone anybody else, and it alarmed me so much that I even considered abandoning my lifelong goal of creating a taxonomy that spanned the entire Indo-Pacific. Then suddenly the fog cleared.
Being a creature of habit, I always get up early in the morning and usually make myself a cup of tea before going to my study. On one such morning in 1993 it was business as usual until, half asleep and waiting for the jug to boil, a concept of evolution very different from Darwin’s came into my head, unprompted yet fully formed. Well, almost fully formed – just how this evolution interfaced with Darwinian evolution took me several years to think through. Revelations such as this are popularly called eureka moments, and occur when a simple solution to a complex problem suddenly occurs to someone who has long pondered the matter, and if the many stories about them are true, they usually do so at unexpected times.
My tea was cold before I recovered. I was angry with myself that I had not worked out before then why differences between a pair of species in one country might not apply in another country. At the same time, I was elated to have found a clean solution to the problems of taxonomy and biogeography that had bugged me since I first worked on corals in Western Australia. These weren’t just my problems: the question ‘What are species?’ had been plaguing taxonomists, evolutionists and many others for as long as I could remember.
I called my concept reticulate evolution, having come across the term in a book called Plant Speciation, written by Verne Grant.26 Grant’s reticulate evolution and mine have plenty in common, but he coined the name to describe something that he thought was peripheral, whereas I believe it’s at the hub of the organisation of Nature, and the main mechanism by which most groups of plants and animals change over time.
The basis of Darwinian evolution – natural selection and survival of the fittest – can be seen in action everywhere and is, I believe, as do all sane scientists, beyond question. However, Darwinian and reticulate evolution are poles apart on practically every point when it comes to when, where and how species come into existence.27 Yet there is overwhelming evidence that both are correct: how can this possibly be?
The very short answer is that Darwinian evolution (and by that I mean neo-Darwinian evolution, the ‘neo-’ bit being genetics, which hadn’t surfaced in Darwin’s time) sits under the umbrella of reticulate evolution. It’s like the icing on the cake, the cake being reticulate evolution.
The slightly longer answer, which focuses on how the term ‘species’ is understood, is that most interpretations of Darwinian evolution are based on the premise that species are reproductively isolated units (that is, they do not hybridise), whereas reticulate evolution is based on the premise that they are not. That initially seems rather facile, but the distinction is fundamental. We humans need units of some kind in order to name, describe or map – to do almost anything that involves communication. If there were another life form able to communicate in terms of continua, our whole concept of how Nature is organised and changes over time would be very different. The crux of the matter is that Nature seldom forms units of any kind below the level of groups of species called syngameons (as I’ll explain below); we must impose units, most commonly via species names, on natural continua because we have no choice in the matter, at least I haven’t thought of one.
What, then, is a species? This simple question has dominated the thoughts of evolutionists for two hundred years.
Mind experiments can help explain the differences between Darwinian and reticulate evolution, ‘evolution’ referring, in both cases, to the process leading to the formation of a new entity that a taxonomist calls a species.
Experiment one: imagine what would happen if all the surface currents in the ocean stopped. Most fish, reptiles and mammals would still be able to move around because they can swim, but most invertebrate species would stay put because they depend on their larvae being moved by currents to disperse. Over a very large amount of time, these isolated animals would gradually evolve (by Darwinian natural selection, leading to survival of the fittest) into separate species, each endemic to the place where it lived. Now imagine the opposite, a situation where currents are so strong and variable that larvae of all descriptions are regularly dispersed everywhere. Under such conditions there would be few species because they would hybridise everywhere and those that existed would have huge distribution ranges.
While neither of these extremes has actually happened, our oceans have oscillated between them, through continental drift, sea level changes, and
random changes in the paths and gyres of ocean surface currents. Changes in ocean currents and the larvae they carry mean changes in the pathways of gene flow, paths that are continually being broken and reforged in time and space. In this scenario, evolutionary change is being created by currents, and currents don’t care about the larvae they carry. This evolution has nothing to do with natural selection, competition, survival of the fittest, or any other biological process.
Experiment two: this one is on land. Today, botanists estimate that there are more than 850 species of what we call eucalypt trees growing in Australia. Imagine their fate if the continent gradually dried out. Over many generations, all would retreat towards the coastlines where they could find enough water to survive. Now imagine the opposite, a geological interval when rainfall is widespread across the continent. The eucalypts would spread out from the coast and gradually populate the whole continent.
These extremes never happened either, but the climate of Australia has oscillated between them for all known time. If eucalypt species were reproductively isolated, these climate changes would simply result in distribution changes, but most are not reproductively isolated, most can interbreed with many other eucalypts, so changes in rainfall patterns create changes in the pathways of gene flow among species, paths that are continually being altered in time and space. As with experiment one, this is evolution but it’s evolution created by environmental changes rather than by natural selection.
Clearly it’s not that simple; how could it be? I’m talking about how Nature is organised and how it changes, one of the most complex subjects imaginable.
Most evenings I can hear laughing kookaburras from Rivendell, beautiful birds that sound and look the same wherever they are, more or less. This is because they can move their genes around their home range – much of Australia – in sufficiently few generations to be able to maintain themselves as a single, genetically cohesive unit, and being a single cohesive unit they can evolve by natural selection. From my study, I also see many poplar gums, eucalypts with distinctive leaves, bark and flowers. However, if I drive away from my house for an hour or so, up or down the coast, these trees start to look slightly different, and if I keep on driving they become very different, until, according to my field guide, they are not poplar gums at all, they are other sorts of gums. What, then, is the range of poplar gums? There isn’t one, or at least not a well-defined one, because they are not units. What, then, is a poplar gum? They are something that was named from pieces of a particular tree growing in a particular place. Every other tree that carries this name is, ever so slightly, not a poplar gum. Poplar gums are units of human creation; they are not entities that actually exist in Nature.
So where do corals fit into all this? They vary a lot with local environment, so we need to revert to imagination again to remove that complication. Some corals are like laughing kookaburras in that they look the same wherever they occur, but most are like poplar gums, although with one huge difference. Most corals live in the Indo-Pacific Ocean, which encircles two-thirds of our planet. They would have to learn to fly if they were to move their genes around that much space to remain a cohesive unit. Yet corals, and many other groups of marine invertebrates, have tackled that problem very successfully – by producing larvae capable of long ocean journeys, not long enough to form a kookaburra-like unit, but long enough to retain some common identity from one country to the next, and so on across their whole distribution range.
Admittedly that sounds like cheating on my part: if I claim that a species known in Fiji also occurs in Madagascar, where it looks rather different, am I not just making two species into one? If at some future date someone tries to interbreed these two and finds they can’t, am I not proven wrong? Well, yes and no. If that ‘species’ also occurs in Chagos, the Philippines and on the Great Barrier Reef, and is a single entity over every part of that range, we may have a continuum. Only where two of these entities can be seen to occur together in the same place (and so have overlapping ranges) do we have separate species.
Is all this not just old-fashioned geographic variation? It appears to be, until it is studied more closely. Unhappily for the student who does this, it just reveals a need for more study, for if the species is not a reproductively isolated unit, then the more it is studied, the deeper the problem gets. I will return to this below, but suffice to conclude here that there is no taxonomic solution to this very basic problem other than to stop imagining that all species are well-defined units.
One question that continually arises is: how prevalent is reticulate evolution? One way of estimating this is to find out which species are reproductively isolated (that is, form units), and which form interlinked groups. I turned to our 2000-odd domesticated plants and animals to try to find an answer and found that almost all can interbreed with other species: almost all are hybrids or could-be hybrids. That includes my dogs, which have a proven pathway of evolution that takes in wolves and even coyotes here and there. In fact I am sure it includes me, for in me the Neanderthals live on. My ancestors didn’t kill them off, at least not all of them; they copulated with them, probably thousands of times, and sometimes the resulting progenies were better off as a result. It’s a happy thought.
I first described reticulate evolution in Corals in Space and Time, the same book that initially revealed the existence of the Coral Triangle, as I will later relate.28 It was published in 1995 by both the University of New South Wales and Cornell University and was reviewed that same year in an article in Science.29 This book gave the Coral Triangle a flying start, but reticulate evolution remained in the doldrums for a long time, such is the enormous inertia of neo-Darwinism. The consequences of reticulate evolution, however, are far-reaching for almost every facet of biology, because evolution must be taken into account if the subject matter has anything to do with species. Only in recent years has it started to come to the attention of other areas of science, often being heralded as a new discovery or a new way of computing data. One day it will be seen for what it is: the central process in the mixing and remixing of genes along the pathways of evolutionary change for most plants and animals.
An important footnote to this subject is that Darwinian evolution is change for improvement – through natural selection – whereas reticulation is just change. The only improvement it might bring about is in the creation of a new genetic entity that Darwinian selection can then work on. The two diagrams below might help you follow this, provided you remember that they represent both time and space, the latter being a third dimension and too hard to draw.
Hypothetical evolutionary change in identical species lineages from an ancestral origin to present time. Left: Darwinian evolution proceeds by divisions of clades through natural selection. Species are mostly defined units which have a time and place of origin. Extinctions occur by termination of lineages. Cladistics should replicate this phylogeny and indicate present-day affinities. Right: Reticulate evolution proceeds by both division and fusion of clades through environmental changes. Species are ill defined and have no time or place of origin. Extinctions occur by both termination and fusion of lineages. Cladistics will not replicate this phylogeny nor give a true representation of present-day affinities.
Another footnote is that Darwin (to whom I often talk, on account of being a little crazy) recognised the fuzziness I speak of and seems happy with reticulate evolution. In fact he thought about it himself, which is why he started gathering information about domesticated plants and animals all those years ago.
We now have computer programs that produce results from the otherwise unmanageable amounts of data that DNA studies produce, and similar programs to do our thinking for us when we analyse morphological data. The results of most types of computer analyses are unfused branches of a would-be evolutionary tree, because the programs cannot allow branches to fuse. Most programs that produce these diagrams, usually called cladograms, do so by a process called cladistics, the brainchild of the German biologist Wi
lli Hennig (1913–1976). That process is supremely logical, provided, as Hennig himself pointed out, there is no gene transfer between the entities being studied: the process must assume that these entities are reproductively isolated.
Cladistics comes into its own on small scales, such as determining affinities of groups within a species – for example, between races of Homo sapiens, because humans cannot interbreed with other species. Cladistics also comes into its own on large scales, such as determining the relationships between the main groups of primates, as these taxa do not interbreed either. However, at intermediate scales, where interbreeding does occur, cladistics can give a false picture of evolutionary lineages that have occurred in Nature.
The diagrams earlier in the chapter illustrate the point. Both sets of imaginary evolutionary pathways, if analysed using cladistics, might give a similar, or even the same, result. That result is true for Darwinian evolution but false for reticulate evolution at the level of species groups. If we call these groups genera, then cladistics may show an incorrect phylogeny.
I have little doubt that this matter will get no more than lip service for many years to come: why find fault with a nice cladogram produced by a computer that readers and editors alike will be happy with? The reality, however, is usually less satisfying; reticulate evolution predicts fuzzy geographic boundaries, fuzzy morphological distinctions, and fuzzy genetic distinctions due to multiple evolutionary pathways. It even says that binomial nomenclature is fuzzy, and that the synonymies that taxonomists love to play with may vary geographically. All very inconvenient, and often confusing, truths.
Curiously, the title of Darwin’s ultra-famous book on the origin of species is almost a misnomer. It is an account of why new species evolve – because of natural selection and survival of the fittest – but it says nothing about the actual mechanism by which species originate. It’s a bit like Newton’s laws of gravity, a quantum leap in physics to be sure, but Newton had nothing to say about the mechanism by which gravity operates, something that initially eluded even Einstein.
A Life Underwater Page 23