Extensive work on the palaeomagnetism of the Deccan traps by several workers has revealed only one reversal (normal-reverse-normal) of the geomagnetic field during the period of eruption of these flows. The scatter in the natural remanent magnetic directions of different flows in a sequence of these traps is rather large and it persists even after magnetic cleaning. Generally this large scatter of directions is attributed to the geomagnetic secular variation during formation of the traps. Recent studies of the magnetic properties of the trap samples indicate that the presence of multidomain magnetite in a lava sequence differs from locality to locality in the Deccan trap province and so the stability of their natural remanent magnetization. While these features can be clearly perceived, an understanding of the same seems to be quite difficult. However, it appears that part of the scatter in directions could be due to the alteration of the magnetic minerals in the Deccan traps and hence it would not be proper to relate the entire scatter to the secular variation.

Similar Sr isotopic ratios (∼ 0.7055) for the tholeiite-spilite flow unit and the associated mineral phases, of Bombay (Deccan Traps) provide a direct evidence for the spilitic degradation of tholeiite. In contrast, a dramatic increase in the rare earth elements (REE) from basalt to spilite is rather puzzling as rare earths are considered to be relatively immobile. The geochemistry thus suggests that the process of spilitization is due to the reaction with a complex fluid having identical Sr-isotopic composition as that of the basaltic magma—thereby masking the details of the mixing process.

We report here a⁴⁰Ar-³⁹Ar age of 66.0 ± 0.9 Ma (2σ) for a reversely magnetised tholeiitic lava flow from the Bhimashankar Formation (Fm.), Giravali Ghat, western Deccan province, India. This age is consistent with the view that the 1.8–2 km thick bottom part of the exposed basalt flow sequence in the Western Ghats was extruded very close to 67.4 Ma.