On pg. 473 fn. 47 in my discussion of ō for au I point out that, although the monophthongized rendering of this sound is typically regarded as "non-urban", there are examples of ō for au from Roman inscriptions of the Republican period, some of which are given in fn. 48. In addition one could add that two old Roman placenames show ō for au. These are Loretum 'laurel grove', a location on the Aventine (Plin. Nat. 15.138: Loretum in Aventino vocatur ubi silva laurus fuit; CIL I 2, p. 240 IN LORETO ) <— laurus and Codeta, a part of the Campus Martius where horsetail grew (Paul. Fest. p. 58: ager trans Tiberim quod in eo virgulta nascuntur ad caudarum equinarum similtudinem) <— cauda. These show that the monophthongal pronunciation was quite well established in Rome itself.
See Bertoldi, Vittorio, 1940. Storia d'un dialettismo nel latino dell'urbe. Rivista di filologia e d'istruzione classica. 18:22–33.