The nature of the accreting companion to Mira—the prototypical pulsating asymptotic giant branch star—has been a matter of debate for more than 25 years. Here, we use a quantitative analysis of the rapid optical brightness variations from this companion, Mira B, which we observed with the Nickel Telescope at Lick Observatory, to show that it is a white dwarf (WD). The amplitude of aperiodic optical variations on timescales of minutes to tens of minutes (≈ 0.2 mag) is consistent with that of accreting WDs in cataclysmic variables on these same timescales. It is significantly greater than that expected from an accreting main-sequence star. With Mira B identified as a WD, its ultraviolet (UV) and optical luminosities, along with constraints on the WD effective temperature from the UV, indicate that it accretes at∼ 10− 10 M☉ yr-1. This accretion rate is lower than that predicted by Bondi–Hoyle theory. The accretion rate is high enough, however, to explain the weak X-ray emission, since the accretion-disk boundary layer around a low-mass WD accreting at this rate is likely to be optically thick and therefore to emit primarily in the far or extreme UV. Furthermore, the finding that Mira B is a WD means that it has experienced, and will continue to experience, nova explosions, roughly every 10 6 years. It also highlights the similarity between Mira AB and other jet-producing symbiotic binaries such as R Aquarii, CH Cygni, and MWC 560, and therefore raises the possibility that Mira B launched the recently discovered bipolar streams from this system.