The steep slopes of the size distributions of the presently known asteroid
families have long represented a debated problem. The reason is that it is
not easy to reproduce them by the usual modeling techniques based on the ap
plication of standard power-laws as suggested by laboratory experiments. In
this paper, we suggest that the failures of the previous models were due t
o the fact that geometric effects were not taken into account. In other wor
ds, the finite sizes of the parent bodies and the fact that fragments tend
to have convex shapes cannot be disregarded. Following this approach, we fi
nd that steep size distributions are necessarily produced by fragmentations
of the parent bodies. Moreover, we have been able to reproduce fairly well
the observed size distributions of the major families, and we have also ob
tained some reasonable constraints on the original sizes of the parent bodi
es. Some anomalous mass depletions, probably due to injection of fragments
in nearby resonances have also been found, not unexpectedly, in a few cases
(Maria, Themis). (C) 1999 Academic Press.