Attribution: You must attribute the work in the manner specified by the author or licensor ( but no in any way that suggests that they endorse you or your use of the work).Noncommercial ‒ you may not use this work for commercial purpose.No Derivative works ‒ You may not alter, transform, or build upon this work.Sharing ‒ Individual scientists are hereby granted permission, without fees or further requests to GSA, to use a single figure, a single table, and/or a brief paragraph of text in other subsequent works and to make unlimited photo copies of items in this journal for noncommercial use in classrooms to further education and science.


Riley et al. (2006a) question our interpretations of 40Ar/39Ar data obtained on the Karoo flood basalts (Jourdan, et al., 2005). We believe their arguments are partially based on (1) poorly reliable 40Ar/39Ar age data (e.g., analyses of whole rocks that might have suffered hardly detectable alteration, recoil, etc.), and (2) mis-interpretation of the reliable geochronological data available on Karoo and other continential flood basalts (CFBs).


Riley et al. argue that the duration of the Karoo province is comparable to some of the other Phanerozoic CFBs that are classically considered to be plume related.

A duration of ~10 m.y. for the Paraná-Etendeka CFB is mainly based on 40Ar/39Ar data obtained by Turner et al. (1994) with laser spot experiments on whole rocks with strong alteration overprint. With this approach, it is impossible to distinguish data affected by alteration or excess 40Ar, and therefore the proposed duration of 137–127 Ma cannot be considered as valid. Instead, the published age data demonstrate that the peak of the magmatic activity is unambiguously at 131–133 Ma (e.g., Renne et al., 1992). The main volume of the Siberian traps was emplaced at 250–251 Ma (cf. compilation of Ivanov et al., 2005). The Deccan traps are mostly emplaced at 65.5 ± 1 Ma (e.g., Hofman et al., 2000). In conclusion, the available robust age data on these CFBs are in agreement with a brief bulk basaltic activity, though much less voluminous dispersed events might have occurred over 5–10 m.y. This feature is completely different from the magmatic activity of the Karoo Province, for which the available selected age data (Jourdan et al., 2005; Jones et al., 2001; Duncan et al., 1997) show the emplacement of large volumes of magmas persisting over ~5–6 m.y. without any unique and dominant peak activity.

Therefore, according to available age data, the Karoo CFB appears different from most of the other provinces which, as mentioned by Riley et al., have been related to mantle plumes. Although our data do not rule out the mantle plume hypothesis, the persistence of large magma production over several million years suggests that a different process may have occurred (e.g., a strong contribution of the lithospheric mantle?).

When Riley et al. refer to their ages at 190 Ma on Dronning Maud Land (including ferropicrites; Riley et al., 2005), we must outline that these data were obtained on whole rocks that are highly sensitive to 39Ar recoil and alteration processes. Riley et al. present strongly disturbed tilde- and/or saddle-shaped (demonstrating the existence of 40Ar excess, recoil process, and alteration) spectra profiles associated with mean square weighted deviate (MSWD) values ranging from 6.5 to 110. We strongly recommend discarding these whole rock ages to correctly define the history of the Karoo province.


Riley et al. question the south to north migration of the magmatism based on one reliable age of the Rooi Rand dike swarm (174 Ma; Jourdan et al., 2004) and other 40Ar/39Ar ages in Antarctica (Zhang et al., 2003; Riley et al., 2005; Duncan et al., 1997).

Our goal was mainly to emphasize that the emission of magma occurred at different places and at different times in the history of the province. Recent plagioclase 40Ar/39Ar ages obtained on dikes from western South Africa seem to suggest that a younger activity (~176 Ma; Riley et al., 2006b) occurred in the south, but include only a few dikes representing a negligible volume. Concerning the Rooi Rand dike swarm, it must be mentioned that these rocks are chemically specific and display an enriched mid-oceanic ridge basalt–like composition, and do not represent a genuine Karoo flood-basaltic event. The only truly reliable ages published concerning Droning Maud Land are five dates (Duncan et al., 1997; Zhang et al., 2003) that are not sufficient to discard our interpretation based on 55 plateau ages in southern Africa


The probability density distribution diagram of Jourdan et al. (2005) shows that there was no brief peak of magmatism for the Karoo province, but several discontinuous eruptions distributed over ~6 m.y.(for the main volume of magma). On the contrary, other CFBs do show an acme of magmatism of ~1 m.y. and are associated with drastic mass extinctions. Compared to these examples, the Karoo shows a lower rate of eruption and might “allow a recovery of atmospheric SO2 and CO2 levels” (Riley et al., 2006a, p. e109), providing an explanation for the absence of a major mass extinction synchronous with the Karoo magmatism.

In conclusion, we agree with Riley et al. that the long duration of the Karoo province, rather than a brief peak of magmatism, cannot by itself rule out a mantle plume origin. However, as the Karoo eruption timing is atypical compared to well-established plume-derived CFBs, a different origin is not so unlikely.