Charlie,
 
I think that you partially answered your own question about significance when
you said that you could think of research questions that can be pursued at
particular late 19th-early 20th century sites.  The key here, for me, is that
each site must be evaluated individually.  As for the other 10,000, as you put
it, each of these may or may not be worth exploring.  Without knowing the
particular contexts (in the broadest sense of the word) I would be hesitant to
argue for or against any of them.
 
I have found (as have other archaeologists before me) that certain theoretical
or methodological approaches to the record can be more (or less) illuminating
than others.  For example, I am finding that late 19th-early 20th century
farmsteads (a dime a dozen in much of eastern North American) pose problems for
those of us interested in conserving the past.  We often lack a theoretical
framework for these sites and/or a methodology to locate and identify the
relatively rare and isolated features (e.g., well, privies, etc.) typically
associated with such sites.
 
Here in southwest Michigan we have been attempting to examine such sites from a
landscape perspective.  The approach is not particularly revolutionary.  We
assert that the built environment was created and reproduced to reinforce
social relations along class, gender, and ethnic lines.  Thus, changes in
landscapes (e.g., fences, gardens, etc.) are likely to inform on changes in
land ownership, economic status, etc.  What we're attempting to do now is
examine some of the diversity that we expect should exist over this period
between sites that have earlier and later components, that represent elite and
working class, rural and urban, and so on.  I suspect that once we've looked at
2 or 3 1870 middle class urban farmsteads (or 5 or 6), then the 6th and 7th
ones may begin to provide us with SOME redundant information, making the
investigation of more of these sites less important (i.e., significant) than
others.
 
I, for one, am pleased that a case for significance can be made for practically
any archaeological site.  This places the burden of proof, so to speak, on the
archaeologist to make the case for significance.  And, of course, the criteria
are constantly changing;  what is significant today may not be so tomorrow.
 
Just some ruminations on the topic.
 
Michael Nassaney
Western Mich Univ
Kalamazoo, MI 49008-5032