And yet another answer!

  All parts from production are potential 10MHz parts (as far as we can
gather, and I've heard no-one from *any* company seriously refute this
principle).  They are tested, possibly individually but more likely by
batch sampling, for performance and stamped accordingly as "4MHz",
"10MHz" or whatever.  It is *possible* that once enough "premium"
devices have been allocated to market requirements, other components of
equal performance are sold at a lesser specification.

  What this means is that there isn't *necessarily* any difference
between a "4MHz" and a "10MHz" one.  Some people test them at higher
clock speeds again and if they feel their testing is adequate to cover
the likely conditions, proceed to use them at the higher speed.

  It is axiomatic that a device which will work at a particular clock
rate will work at *any* slower speed and you will note the PIC is
characterised to DC clocking.  This applies equally to programming.

Cranston Grey wrote:

> JDR sent me 10mhz version instead of 4.

  Which did they charge you for?  You may be lucky!

> My question is: will simple programmers like david Tait's work with
> the 10mhz clock speed.

  The programmer doesn't use the clock; it provides its own.  In fact,
it works better *without* the clock (crystal) connected.

> I have spent some time on the web looking for an anwser, but it's not
> spelled out anywhere.

  It often happens.
--
  Cheers,
        Paul B.