Recent discoveries of small Earth-approaching asteroids by the 0.9 m S
pacewatch telescope (referred to here as S-SEAs) reveal 16 objects whi
ch have diameters similar to 50 m or smaller. Approximately half of th
ese objects lie in a region where few large near-Earth asteroids are f
ound, with perihelia (q) and aphelia (Q) near 1 AU, e < 0.35, and i fr
om 0 degrees to similar to 30 degrees. Possible origins for these obje
cts are examined by tracking the orbital evolution of test bodies from
several possible source regions using an Opik-type Monte Carlo dynami
cal evolution code, modified to include (a) impact disruption, based o
n a map in orbital (a, e, i) space of collision probabilities and mean
impact velocities determined using actual main-belt and near-Earth as
teroid orbits, (b) fragmentation, and (c) observational selection effe
cts. Amor asteroid fragments evolving from low eccentricity Mars-cross
ing orbits beyond the q = 1 AU line provide a reasonable fit to S-SEA
orbital data. Planetary ejecta from Mars is only consistent with low a
nd moderately inclined S-SEA orbits. Asteroidal fragments from the mai
n-belt via the 3:1 or nu(6) chaotic resonance zones rarely achieve low
-e orbits before planetary impacts, comminution, or ejection remove th
em from the system. This source could produce the observed moderate-to
-high eccentricity S-SEAs. Plantary ejecta from the Earth-Moon system
and Venus are only consistent with low-inclination S-SEA orbits. Moreo
ver, constraints set by the planetary cratering record and the meteori
te record suggest that the Earth, Moon, and Venus are unlikely to prov
ide many S-SEAs. All of these results are predicated on the observatio
nal bias computations (Rabinowitz, D.L. 1994. Icarus 111, 364-377) tha
t provide the current definition of the S-SEA population. (C) 1996 Aca
demic Press, Inc.