Bob,
    Thanks for your comments.  To the rest of you readers--clockmakers have indeed learned a great deal
about the issues we've been discussing.  When I wrote an article about a flex-pendulum, online at
http://arxiv.org/ftp/physics/papers/0306/0306088.pdf
  Bob or one of his associates (I've got so much going on right now I haven't taken time to dig out the letter, and 'senior moments' seem to be unavoidable)  pointed out to me that similar studies have been done in the past as part of the effort to improve horology.  As noted in his letter, it would be difficult to come up with an idea that is
truly original.
    This points to the following fact--we collectively (amateur seismology and horology) have something
truly significant to offer the world of seismology.  Anybody who has seriously tinkered with mechanical oscillators knows that friction is the 'name of the game' when it comes to the design of a seismometer.  Because nobody properly understands from theory the friction responsible for instrument limitations, experiment is the key to success.  Faraday and others were not above 'dirtying their hands' to see what worked.  In his case, what proved
successful had nothing to do with 'eyewash' (beautifully crafted but worthless instruments)--rather the focus was
on the 'truth' of the physics.   For those of you who know little about Faraday, I recommend that you read the
following
http://www.rigb.org/heritage/faradaypage.jsp
 If you look at the many discoveries that he made, you'll see that his methods of doing science were not what we
see for the most part today.  In my career I've seen a tragic progression into what can best be described as
"data-taking technicians" who have little practical knowledge of the instruments they use.  Scientists who try,
or even think about building their own instruments are an increasingly rare scientific commodity. This is tragic because the approach departs radically from the methods on which physics was founded.  I have attempted to document these claims in the chapter,  "Building on old foundations with new technologies", part of a book scheduled for release in January, by Nova Science Publishers, titled  "Science education in the
21st century"  .ISBN: 1-60021-951-9
   On separate matters:
Ian,
    your paper-mate-lubriglide pendulum is great!   I am curious, however, as to why you want to use
the messy oil (or any other) damper.  The primary reason for eliminating transient response is for purpose
of conventional data interpretation.  The inability to distinguish between Love and Rayleigh waves (as an example) because of the spherical motion means this is not very important for your system.  Moreover, the period (which I missed if you mentioned it) is I suspect long enough that local noise disturbances are not as important as otherwise might be the case and for which conventional damping helps a lot. To just see earthquakes (for an alarms, as an example) you are likely to be better off eliminating the
damping--partly because dithering by local noises is actually an advantage.  I know that some folks believe that such an approach is tantamount to 'heresy'.  I have already
proven to my own satisfaction that what I'm saying is not ridiculous.  Moreover, I personally believe that strong resistance to the thought suggests some of the blindness that results when we get trapped in the rut of conventional
thinking.  As I heard a theologian once say, a 'rut' is a 'coffin' with the ends kicked out.
    Meredith,
You mention 'standard' crossed rods oscillating up to about 4 times longer than your present arrangement
with magnets.  If your 'standard' setup has no magnets, then I predict that a big part of this difference derives
from eddy current damping due to the induced currents in the steel rods as the oscillate in the strong magnetic
field of your gold-coated units.  This eddy current component is not a 'show-stopper' like the mechnical
internal friction parts.  It is actually advantageous.
  Randall