Blog powered by Typepad

« Duck! here comes Darwin (1) | Main | "For now I am a judge, and a good judge too." »

Saturday, 01 October 2005


Feed You can follow this conversation by subscribing to the comment feed for this post.

Why, for example, do we see absolutely no overlapping between reptiles and mammals?

Because, if such animals existed, we'd have invented a name for them, say "repmals". No doubt Latham would then be asking why there were no overlaps between repmals and mammals. This goes on forever.

For example, consider aquatic mammals: whales, dolphins etc. They are pretty close to an "overlap" between fish (which are types of reptiles) and mammals, aren't they? We choose to call them mammals, but they certainly seem to share a lot of adaptations with aquatic reptiles (I believe that DNA analysis confirms that aquatic mammals evolved from land-based mammals, so we are correct in considering them mammals).

The point is that the way we classify animals is not some abstract thought experiment: it's purely and only based upon *the way the world is*. In fact, it is based upon rather old and incorrect notions about what is a "species" etc. and I'm fairly sure, if we could start again using DNA etc. we'd come up with a different system which would seem less "jumpy" or "discrete". Defining what exactly a "species" constitutes is pretty hard: you know it when you see it comes to mind.

In short, I rather fail to see that Latham has any point here. He's only objection seems to be: "Why don't we see a continual spectrum of animals" to which a simple answer is: "Blind luck: it's just the way the world is, no different from why I can, if I wanted, walk from Paris to Capetown, but not from London to New York: it's just the way the land masses on Earth are!" I'd also say that if, say, we looked at DNA and not phenotypes, we'd see a less "jumpy" or "discrete" picture.

I've found via Google the full text of Dawkins's book online here:

No idea if this is, ahem, legal in copyright terms. The passage which David quotes from is on page 263.

'Matt', I'm off to France tomorrow for a few days so I haven't the time now to respond in detail. Suffice to say that Dawkins insists, absolutely and categorically with no exception, that there are no overlaps and no hybrids - see chapter 12 of 'The Blind Watchmaker'. I can't see why that shouldn't be; when you consider the 'zillions' of living organisms and multiple species, surely there would be either a fossil or a living example of just *one* of these intermediates.

You don't see overlaps because you stopped drawing circles way too soon. Dawkins' point was that we draw bigger circles taking in more smaller circles based on relatedness. At some point we do, indeed, draw a circle that includes chordates with limbs, and that circle includes reptiles, amphibians, and mammals. In fact the Karoo Formation fossils dramatically demonstrate almost each step of the evolution from reptile to mammal, so we DO have a series of smooth transitional fossils which demonstrate exactly the evolution we might expect based on Darwin's observations.

I think you're confusing Dawkins' careful explanation with an apology, which it is not. The evidence is there, and the evidence is clear. Draw more circles, you'll see.

( -- go to "fossil reptiles"; and

We do see "overlaps" between reptiles and mammals - they are the so-called mammal-like reptiles (or, more accurately, non-mammalian synapsids) that are quite well known from the the fossil record. Anybody who goes to the fourth floor of the American Museum of Natural History when they are in Manhattan can see a number of these fossils after they walk out of the halls of dinosaurs.

Matt - fish actually aren't a type of reptile (although the birds are nested within the archosaurs, with a number of clear intermediates between theropod dinosaurs and extant birds preserved in the fossil record). he similarities of whales and dolphins to fish are indeed superficial. There is strong genomic evidence that the whales and dolphins are most closely related to the hippo. The similarities between the fish and cetaceans are the result of convergence (and similar convergence occurred with the now extinct ichthyosaurs, which were reptiles - go to to learn about these organisms, which were fascinating due to their extremely large eyes)

It has been some time since I read "The Blind Watchmaker", but I can state categorically that there *are* many hybrids and intermediates (which are different things). A hybrid is a cross between two different species. Not only has such hybridization led to many changes within the plants (allopolyploids are well established types of plants that result when there is a rare successful cross between related plants that is preserved by maintenance of two sets of chromosomes from each parent - there are even segmental allopolyploids where parts of different chromosomes successfully pair during meiosis and others do not). Even in the animals, drawing some species boundaries can be *very* difficult precisely because there is fairly continuous variation. The isolation of species may be incomplete and due to factors such as behaviors. However, as species diverge additional changes reinforce reproductive isolation.

Finally, I would point out that:

"I need hardly point out also, that no-one has ever seen a new species come into existence despite the very best efforts of scientists experimenting on fruit flies who produce successor generations at a prodigious rate. No matter what they do to them, they remain fruit flies!"

is inaccurate (though I'm sure not intentially so) for several reasons...

1. The purpose of Drosophila genetics has not been to produce a new species.
2. Even if they had *succeeded* in generating a new species, they would not have changed them from fruit flies -- there are many species of fruit fly that are pretty similar to us, but very different in the sense that they cannot successfully cross (different enough that hybrids are actually impossible).

Some of this is simply the perception of the non-expert (to whom a fruit fly is a fruit fly is a fruit fly, with appologies to Stein). However, it is important to stress that the tempo of change in morphology is probably more uneven than Darwin thought. So some changes probably occurred more rapidly than would have been expected. But this is a relatively minor (though fascinating) issue that does not require wholesale revision of Darwin's ideas. Instead, it simply calls out for the incorporation of our knowledge of the molecular basis for development into the neo-Darwinian synthesis (which is ongoing anyway - see Sean Carroll's book "Endless Forms Most Beautiful") just as the archetects of the modern synthesis incorporated Mendelian genetics into Darwin and Wallace's basic framework.

Finally, I would add that the accusation of missing links is inaccurate - whenever a new "missing link" is added it creates two new gaps. Fossils simply weren't formed in numbers that would allow us to fill in *all* gaps. Enough exist - however - to give us confidence in the broad pattern.

Oh, God! ( Well, "Oh, Intelligent Designer", lacks a certain something.)

Sorry, one and all, I've just enjoyed a couple of home-made dry martinis and a decent-ish bottle of Tempranilla, and to be quite frank, I can't cope, can't cope!

I am delighted that we have some experts aboard, and I promise to respond more coherently on my return from France in a few days. I will just quickly say to Edward that I was underwhelmed with his Great Karoo site, whilst admitting that my perusal may not have been of the keenest.

I apologise for the use of the word 'hybrid' which I sort of knew had a technical meaning but I lapsed into laziness. Nevertheless, you all know what I mean, and I still don't understand why there are no intermediaries around now.

Anyway, keep them coming, and until later this week, a toute a l'heure!

'The fact is that if they evolved in a smooth way from a common ancestor then there must have been a time when there was an in-between, transitional form which was neither reptile nor mammal."** He goes on to point out the obvious, that there is no reason why such a hybrid species could not have survived to this day, yet none has.'

They're sterile. Also, what Edward said.

BTW, it would be hybrid phylum, not species. And, contra your reading of Dawkins, cladistics is not as straightforward as you make it seem.

Edward, thanks, I stand corrected on the fish/reptile thing!

Off Topic: Having searched for Latham's book on Amazon, I just went back to and there "recommendations" for me are all anti-Dawkins and pro-ID books. Eurgh!

On Topic: David (and I'll wait a few days, enjoy your France trip!) I don't understand what you mean by "intermediates". There are plenty of examples of different animals living today which have a common ancestor. As people have pointed out here, there are intermediates in the sense of, say, a Lion and Tiger getting it on. What would you expect to see, from evolutionary theory, but don't in practise? Some sort of half-arsed reptile that's becoming a mammal? Well, this is a fascinating Wiki page about the animals Edward mentioned:

They sure sound like my silly idea of a repmal, and these guys sound even more so:

Having re-read some of Dawkins's book online, I note that it's very unfair of Latham to compare Mammals and Reptiles. As Dawkins says:

Mammals are all descended from a common ancestor, which is not an ancestor of any non-mammal. Birds and mammals have a more remote common ancestor, which they share with lots of other animals like snakes and
lizards and tuataras. The animals descended from this common ancestor are all called amniotes. So, birds and mammals are amniotes. 'Reptiles' is not a true taxonomic term, according to cladists, because
it is defined by exception: all amniotes except birds and mammals. In other words, the most recent common ancestor of all 'reptiles' (snakes, turtles, etc.) is also ancestral to some non-'reptiles', namely birds and mammals.

That passage seems to answer Latham's question, doesn't it?

Gentlemen, laziness, induced by soft French living, and a reluctance to kick start my brain is the only explanation for failing to respond to you rcomments - plus, a certain amount of personal life which has the habit of intruding, mostly in the very irate form of the 'little Memsahib' who will not be denied! I shall try and do better over the weekend.

However, I will leave this thought. According to *all* Darwinists, neo or whatever, the process of change from heritable characteristics under pressure of natural selection, is very, VERY gradual. Thus my question remains, how is that Dawkins and others can draw nice, exact (Dawkins insists on it!) tidy rings around each phyla? Where are the remains of the in-betweens?

Each circle is the set of all descendents of a particular individual. The circles must nest because being a descendent is transitive: if X is a descendent of Y, all of X's descendents are also descendents of Y. This is complicated a bit by sex, but not much.

Consider cats and dogs: there's no overlap between the sets of cats and dogs, as they're not interfertile, but they are both contained in a larger set, the order Carnivora. This corresponds to a more distant ancestor common to both.

"Larry (the) Lamb"! My God, that brings back memories. "Pleeeease, Mr. Maaaaaayor, Sir". And what was the name of the daschund?

Sorry, you had to be there!

Anyway, Larry, I understand (just) your mathematical set theory, but I don't think it helps the Darwinian case. The point Latham makes is that Darwinists (with one or three exceptions) insist that evolution occurs very gradually, step-by-step and that natural selection presses on individuals. Consequently, as Latham puts it, "Why, for example, do we see absolutely no over-lapping between reptiles and mammals? The fact is if that they evolved in a smooth way from a common ancestor then there must have been a time when there was an inbetween, transitional form which was neither reptile nor mammal. There is no particular reason why such a transitional form should not actually have survived to today. One should expect to have at least a few such transitional survivors in the whole of the animal kingdom with at least one 'non-perfect nesting'. But we do not and Darwinian theory cannot explain this."

He goes on to point out that 'perfect nesting', in fact, argues against the gradualist Darwinian evolution it is supposed to support (according to Dawkins). he also points out the embarrassing fact, to Darwinists, of an unholy row going on in the discipline(?) of taxonomy. This leads to the rather delightful sight, to me at any rate, of watching Prof. Dawkins fight the creationists with one hand whilst trying to discipline his fellow biologists with the other.

At the risk of tedium, I repeat, in a gradual, step-by-step evolution there should be signs of that very gradualism and not the rigid 'set theory' that Larry describes so well.

David, I've tried to explain why the standard evolutionary model of descent from a common ancestor predicts the "nesting" of taxonomic categories. You say you understand this, then go on to quote an assertion that "Darwinian theory cannot explain this" . As I've just explained (or tried to!) how it does explain exactly this it would be nice if you'd say where you think I'm wrong.

Latham's point about gradualism is ill-founded: changes between individual parents/children are small, but add up to large scale effects over many generations, and the biological diversity we see is the result of many, many, many such generations. Diachronic vs synchronic, if you know those liguistics terms.

Larry, it's late-ish and I'm not at my best. I'm not picking an argument with you, merely struggling to understand.

You are absolutely right, if, but only if, all living things emanated from the first reproducing organism. Using your set theory, anything that is subsequently produced must be part of that original, over-arching set.

But given that, and, just as important, given that Darwinists, not me, insist that all change is very tiny, very gradual, step-by-step, why are there *no* signs, anywhere, of any inbetweens? Why is there, according to Dawkins, a strict demarcation between 'sets' and not a fuzzy blur?

I will provide another example cited by Latham in his book. We, humans, along with all other vertebrates are part of the 'set' called chordates. As Latham puts it, "The origin of chordates is as mysterious as the origin of any known phylum. There are no intermediate transitional fossils to link them to any other invertebrate. Benton[*] writes: 'It is hard to find any reason for pairing adults of the phylum Chordata with any particular group of worms, moluscs, arthropods, or any potential relatives'"
[*]Michael Benton, 1997, "Vertebrate Paleontology".

These gaps are not confined to the animal kingdom. According to Jane Francis, Professor of Palaco-climatology at the University of Leeds, speaking on the BBC earlier this year, "The flowering plants originated about 100 million years ago [...] and we have good fossil there so we can study things in a lot more detail. And [...] we can see the early ancestors of them and we have a very good record. But the interesting thing is we don't see what we think is the very early flowering plants. You know, there's a big gap and we have suspicions there may be 20, 30, 40 millionyears when the flowering plants evolved and we just have no evidence of them".

It is this recurring theme of "no evidence" that intrigues me and the more defensive Darwinists become over it, the more intrigued I become.

No, I was not familiar with the words you used but was, as always (being a curious cove!) happy to look them up, and I can only beg that we do not become diverted into language, a fascinating topic, but not analogous to evolution, I think.

I'd like to reply to the points you raise one by one, to stay focussed. Today's lucky point is: "why are there *no* signs, anywhere, of any inbetweens?".

Let's go back to cats and dogs. Cats and dogs share from a common ancestor in the Carnivora order. This invidual had many descendents, some of which became specialised in hunting particular prey species, others of which went in for being generalists and working in teams. Of course, the former are cats and the latter dogs (well, canids). Note that the steps from the Carnivora ancestor to each of its descendents, mother to daughter, are quite small individually _but_ over hundreds of thousands of generations become significant. There is no overlap between the cats' and dogs' circles; their commonality is expressed by those circles being wholly contained in a larger circle, the descendents of the first member of Carnivora.

(Aside: hyenas, which might be thought of as an "intermediate" twixt cats and dogs are actually cats trying to occupy a dog niche)

Synchrony/diachrony are just ways of talking about the development of things in time, not necessarily restricted to linguistics - but let's forget them for the moment.

The cats/dogs discussion above is illustrative - I'm no expert on this stuff, and may have missed out a lot.

Larry, I suspect we are in danger of going in circles. I recognise your description of the end result and the implied standard Darwinist and neo-Darwinist explanation of that result. My question remains; does it comply with what we actually observe, or to be exact, with what we do *not* observe?

Let me quote form "Chaos and Life" by Richard J. Bird, Columbia Press: "The idea of species is an awkward one for biology in general [...] but it presents particular difficulties for Darwinism. The problem for the Darwinian is that the whole concept of species is really at odds with the idea of gradual change upon which evolution by natural selection depends. If species can change into other species by gradual steps, where do you draw the line between them, and indeed, why have a division at all?"

No-where, as far as I know, does there exist any relics of any living thing that is, so to speak, a half-way house. As a Darwinist, it is for you to offer an explanation for, dare I say it, 'the missing links'!. Before you do, I would add another quotation from Richard Bird's thought-provoking book: "The picture of speciation according to the orthodox view is that many individuals of a species aquire small changes until they collectively form a sub-species that is better adapted. But how does it happen that the favourable mutation occurs in enough individuals at the same time? If it does not, wouldn't inter-breeding between mutants and non-mutants dilute the effects of the mutation?"

Unless anyone else jumps in, I will leave the last word to you as I gird my loins for a new post on another difficulty for Darwinists.

I believe my "cats and dogs" account roughly covers how those animals developed. A fortiori, my claim is that it's _all_ like that, and that this fits what we observe (like the lack of intermediates between dogs and cats) very well.

There's a lot to be said about species, and I suggest you make it a new top-level topic. Happy to say more if you like, but for now I'll just observe that, in biology, the notion of species is "emergent" and that sometimes the line isn't drawn - see "ring species" for example.

Oh no! Not "transitional species"! This _is_ basic. There are lots. Here's a particularly nice sequence:
which goes from shore dwelling quadrupeds to whales in easy stages.

The comments to this entry are closed.