R. S. D'Lemos writes: Bremond d'Ars et al. 1992 propose a model for the Northern Igneous Complex of Guernsey whereby granitic members were generated by melting of the country rocks as a direct result of emplacement of the basic members, and present Rb-Sr data from which they proposed a c. 553 Ma age for the complex. I find these interpretations inconsistent with field, petrographic and isotopic data.
Granite magma generation. The proposal that granitic rocks within the complex have a part crustal derivation is not new (e.g. D'Lemos 1987a; Brown et al. 1990). What Bremond d'Ars et al. propose is that the granite magmas (e.g. Cobo Granite) were formed by country rock melting due to emplacement of basic magma (e.g. St Peter Port Gabbro) into the chamber. With the exception of limited assimilation shown by a small number of gneissic rafts within the gabbro and a c. 6 cm wide zone of melting at the contact between a satellite body and Icart Gneiss (Roach et al. 1991), evidence for country rock melting is conspicuously lacking. Detailed petrographic and geochemical analyses (D'Lemos 1987a) reveal the Cobo Granite to be remarkably homogeneous, devoid of restite blocks or inhomogeneities that could be attributed to local partial melting, and to have a near eutectic (minimum melt) composition (e.g. SiO2 = 76%, n = 15). Textures typical of contact melting (e.g. adjacent to Tertiary centres of NW Scotland) are absent. Contacts between Cobo Granite and Perelle Quartz Diorite (Perelle Gneiss of Bremond d'Ars et al.)