The loss cone: Past, present and future

The capture and subsequent inspiral of compact stellar remnants by central massive black holes is one of the more interesting likely sources of gravitational radiation detectable by LISA. The relevant stellar population includes stellar mass black holes, and possibly intermediate-mass black holes, generally on initially eccentric orbits. Predicted detectable rates of capture are highly uncertain, but may be high enough that source confusion is an issue. Foreground events with relatively high signal-to-noise ratio may provide important tests of general relativity. I review the rate estimates in the literature, and the apparent discrepancy between different authors' estimates, and discuss some of the relevant uncertainties and physical processes. The white dwarf merger rates are uncertain by a factor of few; the neutron star merger rate is completely uncertain and likely to be small; the black-hole merger rate is likely to be dominant for detectable mergers and is uncertain by at least two orders of magnitude, largely due to unknown physical conditions and processes. The primary difference in rate estimates is due to different initial conditions and less directly due to different estimates of key physical processes, assumed in different model scenarios for inspirai and capture.