"Last Rites for Cataclysmic Variables"

There has always been a paradox in our understanding of the evolution of cataclysmic variables (CVs). Angular momentum loss should grind the binaries down to Porb~1.3 hr in 1-3 Gyr, and then drive them apart again on a longer timescale. Most CVs should therefore have undergone "period bounce" a long time ago, and be evolving today towards longer period with light secondaries (<<0.1 Mo). This is apparently contradicted by observation. Where are these old CVs, the expected survivors of period-bounce, hiding? We have found a way to identify them, primarily by measuring the rates of accretion-disk precession – which yield the secondary-star masses. It turns out that the "observational" result was wrong: these *are* indeed the majority species of CVs, but have been missed because of their faintness and reluctance to erupt. These measures of secondary-star mass also yield a precise mass-radius relation for CVs.