Key points are not available for this paper at this time.
We investigate the proposal made by Paczynski in 1987 that the arcs of blue light found recently in two cluster cores are gravitationally lensed elongated images of background galaxies. We show that lenses that are circularly symmetric in projection produce pairs of arcs, in conflict with the observations. However, more realistic asymmetric lenses produce single arcs, which can become as elongated as the observed ones whenever the background galaxy is located on or close to a cusp caustic. We present detailed computer simulations of lensing by clusters using a reasonable model of the mass distribution. Elongated and curved lensed images longer than 10" occur in 12% of the simulated clusters. We conclude that the lensing hypothesis must be taken seriously.
Grossman et al. (Fri,) studied this question.