Molecular Gas in the Powerful Radio Galaxies 3C 31 and 3C 264: Major or Minor Mergers?

We report the detection of $^{12}$CO~($1 \to 0$) and $^{12}$CO~($2 \to 1$)emission from the central regions ($\lesssim 5$--$10 {\rm kpc}$) of the twopowerful radio galaxies 3C~31 and 3C~264. Their individual CO emission exhibitsa double-horned line profile that is characteristic of an inclined rotatingdisk with a central depression at the rising part of its rotation curve. Theinferred disk or ring distributions of the molecular gas is consistent with theobserved presence of dust disks or rings detected optically in the cores ofboth galaxies. For a CO to H$_2$ conversion factor similar to that of ourGalaxy, the corresponding total mass in molecular hydrogen gas is $(1.3 \pm0.2) \times 10^9 {\rm M_{\odot}}$ in 3C~31 and $(0.31 \pm 0.06) \times 10^9{\rm M_{\odot}}$ in 3C~264. Despite their relatively large molecular-gas massesand other peculiarities, both 3C~31 and 3C~264, as well as many other powerfulradio galaxies in the (revised) 3C catalog, are known to lie within thefundamental plane of normal elliptical galaxies. We reason that if their gasoriginates from the mergers of two gas-rich disk galaxies, as has been invokedto explain the molecular gas in other radio galaxies, then both 3C~31 and3C~264 must have merged a long time (a few billion years or more) ago but theirremnant elliptical galaxies only recently (last tens of millions of years orless) become active in radio. Instead, we argue that the cannibalism ofgas-rich galaxies provides a simpler explanation for the origin of moleculargas in the elliptical hosts of radio galaxies. Given the transient nature oftheir observed disturbances, these galaxies probably become active in radiosoon after the accretion event when sufficient molecular gas agglomerates intheir nuclei.Comment: 16 pages, 1 JPEG figure attached, accepted for publication in ApJ