Comment on: Development of spheroid “composite” bombs by welding of juvenile spinning and isotropic droplets inside a mafic “eruption” column by Carracedo Sánchez et al. (2009)

Pelletal and spinning lapilli as well as concentric-shelled lapilli-bombs are present in ultra-alkaline rocks worldwide. There are well-established models for generation of those pyroclasts in diatreme conduit leading to the formation of subvolcanic pyroclastic rocks, i.e. tuffisites, by diatremic fluidisation. In contrast, Carracedo Sanchez et al. (2009) introduced a completely different model for all the above pyroclasts, and composite (concentric-shelled) bombs in particular, in the strombolian deposits of Cabezo Segura II (CS-II), Calatrava, Spain. In their model the pyroclasts are formed in a strombolian eruptive column. On the basis of our observations and physical–chemical constraints, we contend that this model is not feasible. The Authors' (AA) model requires agglutination/accretion by sintering in the eruption column, keeping the particles in a partially molten state. We find that composite bombs from CS-II show evidence for very rapid quenching of the melt preventing sintering, which in any case is impossible in a convective region where the temperature will be reduced below the solidus of the melt. Analysis of terminal velocity indicates that it would be possible to transport pelletal lapilli in a strombolian thrust region, but dense (between 2.25 and 2.85 g/cm3) lapilli and bombs cannot reach a convective region (stage 3 of the AA's model), which in any case is unlikely to exist at CS-II. By contrast, pelletal-, spinning- and concentric-shelled lapilli and composite bombs can be erupted, but their formation does not depend on the surficial extrusion mechanism. The presence of carbonatite at CS-II indicates juvenile CO2 to be fundamental and indicative of diatremic fluidisation. Concentric-shelled bombs composed of agglutinated pelletal- and spinning-lapilli are interpreted as a development in the upper diatreme.