Red clover is valued for its reseeding ability in pastures and hay lands dominated by cool-season grasses.  Our objective was to determine the effect of harvest frequency on the proportion of red clover derived from recruitment from a naturally produced seed bank in an orchardgrass sward.  Plants of red clover were transplanted on 1-m centers in May 1999, and seed was collected in 1999 and 2000.  The area was oversown with orchardgrass in spring 2001, and red clover was allowed to produce seed in late summer.  The area was clipped twice each growing season, 2002-2004, to prevent seed production.  In 2005, 3 harvest frequency (June, June and July, or June, July, and Aug.) treatments were initiated.  Due to severe mid-summer drought, harvest treatments were June, June and Aug., and June, Aug., and Sept. in 2006.  Soil cores for seed bank analysis were collected in July 2006.  The proportion of red clover increased with harvest frequency, with cover estimates in May 2006 of 30, 52, and 47%, for 1-, 2-, and 3-harvest treatments, respectively.  The red clover component of forage production for June 2006 was 49, 74, and 92% for 1-, 2-, and 3-harvest treatments, respectively.  Results of this study indicated the proportion of naturally recruited red clover in the forage increased with increased frequency of defoliation.