In this folder I am putting files describing
some sparse real, skew-symmetric matrices.  For 
each matrix I one text file, compressed
with gzip.

The text file has ancilary date on the first line,
and the rest of the lines are the data from the
non-zero elemetns in just lower-triangular part
of each matrix.  What is listed is row-number,
column number, then a real numbers, separted by tabs.
Theare are printed in decimal format so differ by about
10^-18 from what I was working with in Matlab.
This data starts in the second row.

The top two also has three numbers.  The first
is the number of rows and columns in the matrix.
The second is what I computed as the smallest
singular value of the matrix.  Since the largest
singular value of all these matrices is close to
3.5 the inverse of this gives a good idea of
the condition number.  The third number
in the top row is the computed sign of the Pfaffian,
so either +1 or -1.  For the larger matrices
(W60* and W80*) I did not compute the sign of the
Pfaffian and so put in a 0.0.