Previous
abstract 
Next
abstract
An O or B-type runaway star can be produced either by a supernova explosion in a massive tight binary system (supernova ejection; Blaauw 1961, BAN, 15, 265) or by dynamical ejection from a young star cluster (dynamical ejection; Gies \& Bolton 1986, ApJS, 61, 419), especially from clusters containing many tight binary stars (Leonard \& Duncan 1988, AJ, 96, 222; Leonard \& Duncan 1990, AJ, 99, 608). It has been suggested that the majority of the OB runaway stars must be due to supernova ejection, because the majority seem to have the rapid rotations and anomalously-high helium abundances expected for stars that have been on the receiving end of mass transfer in massive tight binary systems (Blaauw 1993, ASP Conference Series, 35, 207). However, physical stellar collisions can occur during the binary-binary collisions that produce dynamically-ejected runaway stars, and the merged stars that result from such collisions will have rapid rotations and anomalously-high helium abundances. Also, mixing will reset the nuclear clocks of the merged stars to zero, and thus such stars will have nuclear ages that correlate with their kinematic ages, if they are ejected from their parent clusters. New binary-binary scattering experiments carried out by the author that allow for physical stellar collisions and mergers suggest that some of the merged stars do indeed attain speeds large enough to escape from their parent clusters and be classified as runaway stars. Consequently, the predictions of the supernova and dynamical ejection hypotheses are not as distinct as had been previously believed. A very careful observational study of the rotational rates and helium abundances of an unbiased sample of OB runaway stars will be required in order to discriminate between the two hypotheses for their origin.
