Data Set Overview
  =================
    This data set contains the high-inclination asteroid families of
    Gil-Hutton (2006).  A data set of 3652 high-inclination numbered
    asteroids was analyzed to search for dynamical families.  The
    basic data set was the list of 3697 asteroid synthetic proper
    elements taken from the Asteroid Dynamic Site, March 2005 version;
    Knezevic and Milani (2000).  For this analysis, only asteroids
    with sine of proper inclination greater than 0.3 were used.
 
    Three data files are included.  Families.tab contains a list of
    the members of each of the thirteen families, along with their
    proper elements.  Clumps.tab contains a list of the members of
    each of the 22 clumps, along with their proper elements.
    Summary.tab lists the families and clumps, with the number of the
    least-numbered member, the proper elements of the least-numbered
    member, and the range of proper elements for each family and
    clump.  A cluster of objects in proper element space is considered
    a nominal family (clusters related to the breakup of an asteroid)
    or a clump (clusters not necessarily related to breakup events) if
    they have a number of members larger or smaller than a limiting
    number, which is defined as 2.5 times the minimum number of
    objects for a cluster to be considered significant (Zappala et al.
    1995).
 
    Further information about the analysis can be found in Gil-Hutton
    (2006).
 
    Gil-Hutton, R. 2006.  Identification of families among highly
    inclined asteroids.  Icarus 183, 93-100.
 
    Knesevic, Z., and A. Milani 2000.  Synthetic proper elements for
    outer main belt asteroids.  Celest. Mech. 78, 17-46.
 
    Zappala, V., Ph. Bendjoya, A. Cellino, P. Farinella, and C.
    Froeschle 1995.  Asteroid Families: Search of a 12,487-Asteroid
    Sample using Two Different Clustering Techniques. Icarus 116,
    291-314.

Confidence Level Overview
  =========================
    Membership in the families is strongle dependent upon the choice
    of d(0), the cutoff distance.  See the original paper, Gil-Hutton
    (2006) for details of the analysis.