Economies of scale and the fossil record

From: Wesley R. Elsberry (welsberr@inia.cls.org)
Date: Fri Mar 24 2000 - 11:41:56 EST

  • Next message: Steven M. Smith: "Re: Dating Old Rocks (was Dennett's bad word ...)"

    [Quote]

    [...] Ancient rocks clearly preserve less information, on
    average, than more recent rocks. However, if scaled to the
    stratigraphic level of the stage and the taxonomic level of
    the family, the past 540 million years of the fossil record
    provide uniformly good documentation of the life of the past.

    [End Quote - Denton et alia]

    A fossil is some remnant or trace of an individual unit
    (which for colonial species, like corals, may represent
    quite a few individuals).

    I'm going to take a rough metaphor for illustration. Let's
    say that we perceive the findings of paleontology as one
    would perceive a picture from a LANDSAT or other imaging
    satellite.

    We set the scale to "Individuals". That is, after all, the
    actual scale of the fossils themselves. The picture that
    pops up is a very large but almost entirely blank sheet.
    With a microscope, one might here and there rarely find
    a non-blank pixel.

    That wasn't very satisfying, so we reluctantly move to a
    lower resolution, that of "Species". Immediately, we are
    able to see some scattered dots, about one non-blank dot
    for every 50,000 or so blank ones, but still enough to see
    some clustering and interrelationships. But if we split
    up our data according to time, the quality of the picture
    changes over time, becoming better with more recent data.
    What we want is to see the same quality of picture over
    time at some resolution. "Species" isn't it.

    OK, drop another notch in resolution, and flip through the
    images over time at "Genera". The blank space becomes
    less overwhelming, but we still have the annoying change
    in quality over time. We'll move on.

    Now our resolution is set to "Family", and at least for
    the most recent 540 million years we see no great changes
    in quality as we flip through the images. Those annoying
    blank spaces are somewhat reduced, too.

    What are the properties of our low-resolution images at the
    "Family" setting, though? Each pixel or visible element
    represents a family. For each pixel or visible element, this
    means that there might be as little as one individual shell,
    tooth, or other trace to instantiate that family. There
    could be as much as a whole strata of fossiliferous
    monoculture, as in some chalk deposits, but that still gets
    reduced to one pixel, just like a single tooth could produce.

    Consider the order of primates, comprising a few families. A
    minimal set of "uniformly good documentation" under Denton et
    alia would be one tooth per family per stage, and this set of
    data could easily be the prize in a jumbo box of Cracker
    Jacks.

    Our museums and universities may be able to clear a lot of
    storage space if they take up Denton et alia's scheme as
    a prescription for what should be kept. The compression
    ratio that could be achieved puts to shame those typically
    found in image processing.

    I read the "uniformly good documentation" phrasing with
    emphasis on the "uniformly", not on the "good". They could
    have, with equal justice, have stated their results as
    showing that at the level of family and stage that one
    finds "a uniform level of documentation".

    Wesley



    This archive was generated by hypermail 2b29 : Fri Mar 24 2000 - 11:07:43 EST