Time-delay Measurement for the Quadruple Lens Rx J1131−1231

We have measured the three time delays for the quadruple gravitational lens RX J1131−1231 using two seasons of monitoring data. The short delays between cusp images are A–B = 11.98 −1.27 days and A–C = 9.61 −1.57 days. The long A–D delay for the counter image is not as precisely determined because of the season gaps, but the data suggest a delay of −87±8 days. The short delays are difficult to explain using standard isothermal halo models of the lensing potential, which instead prefer A–B and A–C delays of ∼ 1 day for reasonable values of the Hubble constant. Matching the cusp delays is possible by adding a significant (∼ 5× 10 M⊙) amount of matter nearly coincident (∼ 0. 05 SouthEast) with the A image. Adding such a satellite also helps improve the quasar and lens astrometry of the model, reduces the velocity dispersion of the main lens and shifts it closer to the Fundamental Plane. This is suggestive of a satellite galaxy to the primary lens, but its expected luminosity and proximity to both image A and the system’s bright Einstein ring make visual identification impossible even with the existing HST data. We also find evidence for significant structure along the line of sight toward the lens. Archival Chandra observations show two nearby regions of extended X-ray emission, each with bolometric X-ray luminosities of 2-3 × 10 ergs s. The brighter region is located ∆θ ≈ 153 from the lens and centered on a z = 0.1 foreground cD galaxy, and the fainter and presumably more distant region is 4-5 times closer (in angular separation) to the lens and likely corresponds to the weaker of two galaxy red sequences (which includes the lens galaxy) previously detected at optical wavelengths. Subject headings: gravitational lensing: individual (RX J1131−1231)