S&L+ Where Do We Go From Here?

From
Date
Reply-To
I'm continually amazed by the consensus we're reaching here.  A few
responses to yesterday's postings (in no particular order)

Peter Sheehan paraphrases Bob Kleinpell as follows, ""boundaries must be
drawn with a wide brush, not a finely pointed pen." And this is because of
what we now call the Signor-Lipps Effect."  I couldn't agree more though
Hollis Hedberg pointed out that biozones are not equivalent to chronozones
long before Kleinpell (and even Hedberg was not the first to realize this).
However, the distinction between biozones and chronozones exists
irrespective of the S&L Effect, though this phenomenon certainly
contributes to that distinction. Perhaps everyone on PaleoNet recognizes
the distinction but I keep running across papers in the literature in which
the biozones are routinely interpreted to have time significance down to
the smallest possible scales.  That was what I was trying to get at by
saying that recognition of the S&L Effect should encourage us to keep the
distinction between biozones and chronozones in mind at all times and to
make sure our students do the same.

John Alroy's comment seems to assume that I was referring to Shaw's (1964)
description of graphic correlation.  This was not the case.  There is much
more in that book than graphic correlation and I would encourage everyone
on PaleoNet to take a close look at it.  So far as I am aware Shaw was the
first to ask the question, "How can I determine whether the observed last
appearance of a fossil constitutes a reliable indication of its true last
appearance?"  In doing that, and attempting to answer the question
statistically, Shaw started people thinking about the reliability of last
occurrences, stratigraphic confidence intervals and the rest.  The
conceptual link to Shaw's work is explicitly discussed by C.R.C. Paul
(1982) which is what most people cite as the first paper on stratigraphic
confidence intervals.  The link between the S&L Effect and stratigraphic
confidence intervals is obvious.  [Note: I wasn't seriously suggesting that
the S&L Effect be renamed the Shaw Effect, just emphasizing it's
intellectual heritage.  Besides, I like Stefan's "Tom & Jere Effect" much
better.]

I'd be happy to calculate the stratigraphic confidence intervals for the
Hell Creek vertebrates.  I'm just guessing as to what the answer might be.
The trouble is I don't have the data.  If someone wants to place the data
in the PaleoNet FTP Site, then we can all do it.  That would be fun.

Finally, Alroy proposes to look for overlaps in fossil ranges as a test of
the catastrophic scenario.  I like this because the K/T planktic
foraminiferal record shows more overlaps than disjunctions across the
boundary horizon if you eliminate the spurious signal coming in from
demonstrably incomplete successions.  But I'm not so optimistic about
applying it to the terrestrial record.  This can only be done if there is
some reason to believe that the K/T terrestrial record provides an
opportunity to sample temporally complete (ideally over intervals that
match the event's estimated duration) sequences from many different
environments / geographic localities.  So far as I know (please correct me
if I'm wrong) the only upper Maastrichtian / lower Paleocene terrestrial
faunas of which we are aware come from two localities (one in Montana the
other in China) that represent two different environments.  I agree with
Peter that it's doubtful the terrestrial K/T record is up to that kind of
stratigraphically-oriented test.

To sum up, I see the Signor-Lipps Effect (there, I've said it) as being of
a piece with a new level of appreciation among some of us for the true
nature of our data.  Other strains of this heretical virus include work on
stratigraphic confidence intervals, quantitative biostratigraphy
(especially graphic correlation which IS my own personal favorite), and
quantitative biogeography.  Many of these ideas and the techniques needed
to implement them have been around for quite some time.  [Note: at our
recent graphical correlation research conference Keith Mann brought down
the house when he pointed out that 10m of new oceanic crust has been
created in the East Pacific since Shaw published in 1964 book!!!]  Yet I
don't think that you could describe any of these methods as being routinely
employed or, to go back to the S&L Effect, routinely taken into
consideration.  I think (hope) that it's time to raise our field's
awareness of these issues/techniques and, even more importantly, our
standards with respect to what constitutes a valid
stratigraphic/biogeographic inference.  Since we all seem to agree about
the general importance of these insights and (I hope) the value of
parsimony and formal hypothesis testing (as John Alroy nicely pointed out),
how do we go about moving our colleagues along in this general direction?


Norm MacLeod


----------------------------------------------------------------------------
Norman MacLeod
Senior Research Fellow
N.MacLeod@nhm.ac.uk (Internet)
N.MacLeod@uk.ac.nhm (Janet)

Address: Dept. of Palaeontology, The Natural History Museum,
                     Cromwell Road, London, SW7 5BD

Office Phone: 071-938-9006
Dept. FAX:  071-938-9277
----------------------------------------------------------------------------