This paper shows how 226Ra– 230Th disequilibria can be used to date Holocene volcanic rocks from some well selected volcanoes. A systematic study of these disequilibria on historical or well-dated volcanic samples is indeed first required to test the applicability of this method. Two examples are described here to illustrate its potential. In the case of Mt. Etna, the good correlation observed between ( 226Ra) 0 activities at the time of eruption and Th contents in lava flows from the last two millennia [M. Condomines, J.C. Tanguy, V. Michaud, Magma dynamics at Mt. Etna: constraints from U–Th–Ra–Pb radioactive disequilibria and Sr isotopes in historical lavas, Earth Planet. Sci. Lett. 132 (1995) 25–41] is used to infer the ages of several newly analysed lava flows. The calculated ages are in good agreement with those deduced from the archaeomagnetic curve describing the variation of the geomagnetic field direction in southern Italy [J.C. Tanguy, I. Bucur, J.F.C. Thompson, Geomagnetic secular variation in Sicily and revised ages of historic lavas from Mt. Etna, Nature 318 (1985) 453–455, J.C. Tanguy, M. Le Goff, V. Chillemi, A. Paiotti, C. Principe, S. La Delfa, G. Patane, Variation séculaire de la direction du champ géomagnétique enregistrée par les laves de l'Etna et du Vésuve pendant les deux derniers millénaires, C. R. Acad. Sci. Paris 329 (1999) 557–564, J.C. Tanguy, M. Le Goff, C. Principe, S. Arrighi, V. Chillemi, A. Paiotti, S. La Delfa, G. Patane, Archaeomagnetic dating of Mediterranean volcanics of the last 2100 years: validity and limits. Earth Planet. Sci. Lett. 211 (2003) 111–124]. We also present a whole set of new U-series data on historical, recent, and older samples from Merapi (Indonesia), and show that the ( 226Ra)/Ba ratio has probably maintained a quasi-steady state value during at least the past four millennia, and can be used to infer the ( 226Ra) 0/Ba ratio of old volcanics at the time of eruption, and thus their ages. Comparison with 14C ages available on three samples [R. Gertisser, J. Keller, Temporal variations in magma composition at Merapi volcano (Central Java, Indonesia): magmatic cycles during the past 2000 years of explosive activity, J. Volcanol. Geotherm. Res. 123 (2003) 1–23] shows an excellent agreement. These dating methods, based on the post-eruptive decrease of 226Ra excesses can be confidently used to date young rocks on both volcanoes, an important step to infer their recent eruptive history and magmatic evolution. It also opens the possibility to extend the geomagnetic field variation curve back into the past few millennia. The promising results obtained in this work should encourage new systematic U-series studies to test the applicability of such methods to other permanently active volcanoes showing 226Ra excesses.