Jim's "precise" account nearly squares with a rough equivalent I use of
multiplying late 19th century numbers by 20.  But any such equivalent
needs to be taken with a heavy dose of salt.  Taxation then was much
lower; wage, health, and safety
protections and costs were minimal; costs of many staples (eg clothing)
relatively much higher; automobiles nonexistant; fresh fruits and
vegetables in the winter were luxuries for the rich--in short, many of the
costs of living now used to calculate the CPI were nonexistant or
radically different.  And
because the federal reserve system is
a twentieth century invention, currency is not equivalent either--wildcat
banks issued their own notes, usually heavily discounted in trade; some
companies
payed in scrip--worthless anywhere but at a company store.  And it was a
period of steady deflation and tight money, so it was more difficult to
borrow.  Thus, many things that still exist in the CPI, such as the cost
of money itself, are not really equivalent. Across the board, the kinds of
things included in the CPI
have changed so much that equivalents are difficult things to calculate.

        Another way to calculate the relative value is to look at what
consituted a "living wage" by an average day's pay.  But the much larger
gap between poor and rich, between wage laborers and others, gives a
different number.  Assuming that a day laborer's daily wage is the
standard (and this is not accounting for the fact that most industrial
workers did not work regularly over the course of a typical year, but when
they did work, they worked 12 hour days, six days a week), then one needs
to multiply an 1894 number by at least 25 to get a 1998 number.  So that
$200,000 debt of Twain's looks like anywhere from 3.5 million to 5+
million depending on the standard.

Gregg Camfield