“…Therefore relatively high pressure melting of carbonated peridotite could play a role in generating the CaO‐rich, silica‐undersaturated melts identified in this study [ Della‐Pasqua and Varne , 1997]. However, we consider this unlikely based on the significant differences between the CaO‐rich, silica‐undersaturated arc magmas identified in this study and both experimental melts obtained at high pressures for natural and simple carbonated peridotite systems [ Dalton and Presnall , 1998; Hirose , 1997b] and primitive melilitites, nephelinites, and related rocks from oceanic, continental and arc regions [ Alibert et al , 1983; Dupuy et al , 1989; Wedephol et al , 1994; Wilson et al , 1995; Maaloe et al , 1992; Hoernle and Schmincke , 1993; Clague and Frey , 1982; Cheng et al , 1993; Tatsumi et al , 1999], which are widely considered to be melts of carbonated peridotite [ Eggler , 1978; Huang and Wyllie , 1974; Eggler , 1974; Adam , 1988; Hirose , 1997b; Brey and Green , 1975; Brey , 1978; Wallace and Green , 1988]. For example, the CaO‐rich, silica‐undersaturated arc magmas identified here have higher SiO 2 and Al 2 O 3 and lower MgO contents at a given CaO content than do the experimental melts obtained at high pressures for natural and simple carbonated peridotite systems (Figure 7).…”