[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Subject Index][Author Index]

Life Beyond the Cladogram, episode 2 [ L O N G ].

Thanks, Paul, for a splendid response [Thu, 24 Dec 1998 09:17:47 - Re: Life
Beyond the Cladogram - From: "Dr Paul Willis" <pwillis@ozemail.com.au>]..
It couldn't have been better if it were scripted - a fine demonstration of
the point I was trying to make.

To reiterate that point:  evolution, phylogeny, cladistics are all valid and
productive areas of scientific enterprise.  I'm not taking issue with any of
them (here).  My point is that those subjects tell us little, or nothing, of
value when we seek to explore the 'natural history' of extinct organisms.
I'm definitely not - repeat NOT - trying to provoke another War of the
Clades.  Let me take your responses point by point.

1.  "... my original post did appear to be more dismissive of alternatives
than it ought to have been...".
    You simply stated, emphatically, what you take to be an important fact
of our science - namely, the central importance of phylogeny.  Many others,
and probably the vast majority, will agree with you.  (After all, no one
objected to your statement.)   If you accord such great importance to
phylogeny, then it certainly deserves the emphasis.

2.  "...there is merit in other approaches to palaeobiology than
understanding phylogeny ...".
     At risk of jumping the gun, I'll remark that understanding phylogeny
tells us little or nothing of palaeobiology, whereas those 'other
approaches' are far more profitable.  (I'll try to justify all that in a

3.  "I'd still hold that understanding the phylogeny of an organism places
major constraints on our understanding of its palaeobiology."
     You can gaze forever at a cladogram, and it won't tell you anything
worthwhile about palaeobiology (in the sense of 'natural history').  The
cladogram merely provides a (provisional) phylogenetic bracket.  In a
cladogram of extant tetrapods, dinosaurs would be bracketed between
crocodilians (less derived) and birds (more derived).  From that bracketing
you can infer that characteristics shared by crocodilians and birds were
probably present in dinosaurs as well (at least in generalised form, and at
least in the most generalised dinosaurs).  Thus, you can infer that
dinosaurs had nucleated erythrocytes, asymmetrical gonads within the
visceral cavity, eversible phallus, cleidoic egg with mineralised shell...
and a whole list of other features shared by living crocs and living birds.
(Of course, these generalised attributes are likely to have been overwritten
and remodelled by unique modifications among more derived dinosaurs, just as
they were overwritten by unique modifications in crocs, on one hand, and
birds on the other.)   From the phylogenetic bracket you can derive a
somewhat sketchy broad-brush picture of dinosaurian biology. Now, that is
the BEST picture you can ever hope to obtain from phylogeny pure and simple.
And even that sketchy picture is obtainable only because the phylogenetic
bracket happens to be defined by two extant groups (where you can examine
things like blood cells and reproductive organs).  If the cladogram
comprised only extinct forms, then it could tell us no more than we put into
it in the first place. (Thus, if crocs and birds were both totally extinct,
all we could ascertain about dinosaur biology would be a list of skeletal
and dental characters. )  
      Does such a sketchy picture really impose "major constraints on our
understanding of its [the group's] palaeobiology".   In literal terms, I'm
inclined to agree.  An obligation to start from the phylogenetic picture
(such as a cladogram) imposes a major handicap on our understanding of
     More seriously, an interesting question lurks behind Paul's words...
and I suspect that it's also lurking at the back of his mind.   Do the
generalisations derived from the phylogenetic bracket really represent
BIOLOGICAL constraints?  Reading between the lines, I think this is what
Paul is driving at:  that the natural history of birds (on one hand) and
crocs (on the other) define a sort of 'biological ball-park' within which
dinosaurs must have lived out their lives.  And, by looking at a picture of
phylogeny (such as a cladogram), we can more or less map out the geography
and the boundaries of that 'ball-park'.  While I wouldn't disagree with
this, I would emphasise several points.
     (A) That 'ball-park' is very roughly delineated.  Phylogeny provides a
crude sketch map, not a detailed tourist-guide.
     (B) The ball-park is so big that it may tell us very little of
importance.  It's rather as if someone asked "What's it's like in
Beijing?"... and, looking at a crude map, you can tell them "Well, it's in
China... probably."   In terms of dinosaurs, someone asks "How did T. rex
reproduce?"... and, looking at the phylogenetic bracket, you respond "Well,
it laid eggs... probably".  True enough, but this doesn't tell you anything
particularly special about T. rex, because the same answer will apply to the
whole phylogenetic bracket (all crocs, all birds, and all dinosaurs), not
just T. rex.  To provide better answers, we need to abandon the cladogram
(or map) and go elsewhere in search of information.
     (C) We just don't know if the boundaries of the 'ball-park' are
insurmountable barriers (real biological constraints that dinosaurs could
never break through) or just fortuitous painted lines (that dinosaurs could,
and did, cross - even though living crocs and living birds happen to remain
within the boundaries).  Here understanding of biology is so deficient that
it's very difficult, often impossible, to reach a decision.  No extant crocs
and no extant birds are viviparous... but does that fact betray the
existence of some genuine biological constraint that prevented all crocs,
all birds and all dinosaurs from ever achieving viviparity?  Or is is
possible that no such constraint exists... and that living crocs and living
birds, for reasons unknown, have simply declined the option of achieving
viviparity?   We just don't know for sure.  If we don't know, for sure, that
a constraint even exists, we can't impose it on our investigations of
dinosaurian biology.

4  "If we didn't have mammals larger than half a tonne, we would be able to
construct more about their biology knowing that they are related to humans,
mice and bats than by constructing hypotheses about their biology through
comparisons with other extinct large animals." .  Also, "phylogeny would
seem to me to offer a greater range of insights into the palaeobiology of an
     My original post didn't concern the merits of phylogeny versus
"comparisons with other extinct.. animals".   It identified THREE major
approaches into palaeobiology  - (1) trace fossils, (2) reductionist
programs and (3) analyses of functional anatomy.  Of these, only the third
is likely to include "comparisons with other extinct.. animals".  (Even
then, such comparisons would be a minor pursuit, not a major theme.)   If
you allow me to use all the non-phylogenetic approaches that I mentioned,
these will offer a much 'greater range of insights' than will phylogeny.
I'll take up your example.
     Imagine that there are no large mammals alive today, and then we
discover a real giant in the fossil record (2 tonnes!).  Looking at the
phylogenetic information (cladogram or evolutionary tree of mammals), what
can we ascertain of this creature's natural history?   It's bracketed among
the mammals, so it's (probably) warm-blooded, hairy or furry, viviparous,
with metanephric kidneys, diphyodonty, heterodont dentition... and all the
rest of it.  This doesn't tell us anything of real interest.  It's simply a
description of the mammalian 'ball-park' wherein all mammals exist, and not
just our new monster from the fossil record.   Alright, you can narrow down
the phylogenetic bracket - it's an artiodactyl, with hooves, multi-chambered
stomach... and all the rest of that.  All we get here is a description of
the artiodactyl paddock within the mammalian ball-park.  The phylogenetic
context provides a description that applies to ALL artiodactyls, not just to
our extinct giant.   Narrow it down further... it's a bovid, a titanic cow.
All extant bovids are less than 2 feet high, but this one's enormous. The
detailed phylogenetic analysis tells us that it's closest relative is the
Scandinavian Forest-Cow, Bos horrificus...  So?  What have we discovered
that's special about the biology of our monstrous fossil mammal?   What CAN
we discover about its natural history by studying its phylogenetic context?
Very very little.
     Throw away the cladogram, and (a) look for some trace fossils, (b)
construct some reductionist models, and (c) search out some analogies.
    Trace fossils?  It's got teeth-marks all over its snout, and the
teeth-marks match its own incisor teeth in number, size, shape and
arrangement.  Obviously it got into a fight with one of its conspecifics.
It seems to have been a vicious brute, perhaps like a crocodile or a
vice-chancellor, and quite unlike the placid Scandinavian Forest-Cow..  Rare
and solitary trackways of gigantic bovids, with footprints of appropriate
size and shape, occur in swamp deposits of appropriate age -  though the
skeletal remains were found in alluvial sediments, showing obvious signs of
post-mortem transport.
   Reduction?  Measure up the jaw bones, restore the size and position of
principal jaw muscles, perform a few calculations (or even build a working
model) - and we can start to derive some ideas about bite force, jaw
movements and feeding.  Are these notions testable?   Yes, check the teeth
for predicted patterns of tooth wear.  We now have a hypothesis of how the
creature fed, and we can start to predict the physical properties of the
food material (i.e. what sort of food it consumed).
     Analogies?  it's got no horns, but it does have grotesquely enlarged
canine tusks curling up in front of its eyes.  Nothing similar exists in any
other bovid, living or extinct, but there are comparable tusks in some
living pigs (babirussa) and rodents (Catalonian Corkscrew-Rat).  The
function(s) of those weird tusks can reasonably be inferred by observing
their special functions in pigs and rodents - but NOT in the closest
relatives (other bovids).  Here we obtain palaeobiological information only
by disregarding the phylogeny.
    Using those sorts of approaches (and others)  we could construct a
detailed and specific picture of the animal's natural history:   Many
components of that picture would generate testable (falsifiable)
predictions, and NONE of them would be derived from phylogenetic analysis.
Here, for example, there's evidence to suggest that the beast was
aggressive, solitary rather than gregarious, and lived in swamps.  How on
earth can you get that sort of 'natural history' information from
phylogenetic analysis?
     In short, phylogeny provides a crude 'ball-park' impression of the
animal's palaeobiology (natural history).  .But the details, unique to that
specific animal, don't come from phylogeny (our knowledge that it's "related
to humans, mice and bats").

So, Paul, I have to disagree when I hear, yet again, the pervasive idea that
phylogeny is of central importance in palaeobiology.  Let me repeat that I'm
NOT taking a swipe at you (or anyone else), nor at evolution, phylogeny.
cladistics.  Your original post just happened to provide an irresistible
opening, and your second post is a fine example of a fighting defence in the
cause of phylogeny!

Have a good break over Christmas & New Year.  See you at CAVEPS (?), where
I'll try to placate you with one or two high-octane beverages.  Alright,
three or four...

Back to the swamp....

Tony Thulborn