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

Re: the cause(s) of MASS extinction

On Thu, 13 Aug 1998, David L. Martin wrote:

> 1)  "Dinosaurs were always fluctuating in diversity and were already in
> decline at this time." - Before we can talk about what caused this mass
> extinction, we have to AGREE that it was a mass extinction, which is to
> say, we have to agree that it is something EXCEPTIONAL and not an
> everymillienum, run-of-the-mill event.  Once we agree on this point,
> presumably we can argue about whether this exceptional event has
> exceptional CAUSE(S).

And I claim the only thing that made it exceptional was the extinction of
non-avian dinosaurs.  And we _do_ need an exceptional cause for this.  But
I don't believe the bolide provides one.

> 2) "Just because an impact and an extinction event occurred together
> doesn't mean one caused the other." - I really don't see how this
> contributes to a scientific discussion of this question.  We can say the
> same about any question concerning dinosaurs.  If someone has live
> dinosaurs to do controlled experiments on, please let me know.  Otherwise
> all our inferences about causes will be based on correlations.  Give me an
> alternative or give me a break.

All I am saying is that it is not an open and shut case.  Others would
have it differently.  Practically all the world now takes it as received
wisdom that non-avian dinosaurs were wiped out by a bolide.  But this
explanation is weak because it does not explain the pattern of extinctions
_nor_ their tempo!  The extinctions do not appear to resolve down to a
moment in time.  Many "victims" were already gone, many victims were
already rare, and many victims already have substantial alabis (for 
example, since in some places placental extinction was zero and marsupial
extinction significant, competition is, I believe, a more parsimonious
explanation.  This also works for enantiornithines vs. neornithines!).
Subtract all these (and others!) and you don't have a mass extinction.
> 3) "Lots of impacts occurred before without causing such extinctions.
> Therefore impact cannot be the cause of this one." - This is reminiscent of
> people who live in coastal areas, who experience several hurricanes and
> lose their fear of them.  It takes much more than a powerful hurricane to
> produce truly dramatic devastation.  The storm must be moving perpendicular
> to the coast, have high translational velocity, strike at high tide, AND be
> powerful.  With this combination of factors, a dramatic storm surge will
> produce very effects far greater in intensity and coverage than any
> hurricane winds no matter how powerful.  I could enumerate any number of
> factors that would modify the effects of the impact of a given object.
> Others have already done so.

I applaud this recognition of the likely complex interactions working at
the K/T.  Usually we hear of planet-ravaging fire-storms, boulders
launched into orbit, acid rain drenching etc., etc.  

> 4)  "Why should a global catastrophe kill all the dinosaurs, big and small,
> and not all the birds, mammals, crocodilians, frogs, insects, spiders...?"
> Tally up all the species that are known to have gone extinct in the last
> 100 years.  Look at their size distribution.  Notice anything?  When there
> is a general environmental stress, large species are generally the first to
> go.  Nevertheless, there is a stochastic component.  There is no magical
> body size above which you are doomed and below which you are saved.

Yes.  This is by no means a universal rule.  It depends perhaps more on
fecundity than size.  Afterall, a presumed primary advantage of large
size is to _protect_ you from environmental vicissitudes.  Only
inasmuch as big things tend to be less fecund than small is it likely to
be true.  But non-avian dinosaurs were extremely fecund for their size.
By the time one elephant had one baby (20 months?) a hadrosaur could have
had more than twenty!  This is one of the great mysteries.  They _should_
have been able to recover pretty quickly unless 1. They all went out at
the same time.  or 2. They were targetted by something unrelated to the
bolide; something that was persistent.

> Dinosaurs were around for 180 million years. They had seen many 
> environmental fluctuations, dramatic floristic changes, and the
> introduction of major new animal taxa into their world.  They weathered all
> of these, yet disappeared QUITE ABRUPTLY at the K/T boundary.

Not known.  There are only a couple of places with a good transitional
record.  And, notwithstanding rumours of conflicting data, the closest
analyses suggest a gradual extinction, not an abrupt one.

> Pterosaurs also disappeared, ABRUPTLY. 

Not true.  Pterry diversity was declining to the point where only one?
species is known to have made it close to the boundary.

 Lots of other things disappeared, AT THE SAME
> TIME.  A large impact occurred, leaving an EXCEPTIONAL iridium signature,

Not true.  As I have said before, the closer you look, the further apart
are these extinctions.
> It is for each of us to draw our own conclusions.  Like everything in
> science, it is not about "proof."  It is about working hypotheses.  Saying
> an impact caused the dinosaur extinction is no different than saying
> cigarette smoking causes lung cancer in humans.

Not at all.  People who smoke get lung cancer at a higher rate than those
that do not.  We see them smoking.  We can rule out other factors by
relatively direct observation.  This is more than circumstantial.
There is _NOTHING_ other than circumstantial evidence linking dinosaurs to
the bolide.  And there is a good reason why circumstantial evidence is
viewed with suspicion both in the court and scientific circles--it is
often misleading!

>  It does not mean it is the
> only cause, nor is it an oversimplification of a complex process, nor is it
> a certainty.  It is merely a working hypothesis that you may take or leave.
>  We cannot do controlled experiments on humans to "resolve" the smoking
> question, and we cannot do controlled experiments on dinosaurs to "resolve"
> the extinction question.

I'm not sure this is a very good example.  The smoking question is

> Even if we could, it would not constitute
> "proof," since ultimately we would be making inductive inferences and
> taking measurements and deriving probabilities of type 1 and type 2 error
> and doing all those little things that should remind us that science is not
> about certainty.