Kimberlites intrude the late Archaean eastern Dharwar Craton in two spatially separate fields ( 150 km apart, near Mahbubnagar in the north and Anantapur in the south) to the west of the Proterozoic Cuddapah Basin in southern India. Two lamproite bodies intrude the thick Cuddapah Basin sediments near Chelima and Zangamarajupalle, and a third occurs just outside the present NE margin of the basin near Ramannapeta. Chalapathi Rao et al. (2004) have studied this spatially, and possibly temporally, close association of kimberlites and lamproites to provide insights into their genesis and genetic relationship. Although the main emphasis of their paper is the petrochemical characterization of these two suites of rocks, they also report Sr–Nd isotope data for a small subset of samples, which they use to provide constraints on the nature of the mantle source regions. Of some 65 samples analysed for their whole-rock composition and mineral chemistry, only eight are from the three lamproite bodies. Possibly as a result of their intrusion into the thick Cuddapah Basin sediments, the Chelima and Zangamarajupalle lamproites have high LOI (8–16%) and high values of the contamination index (2 4–3 5), and contain extensive secondary carbonate. This clearly compromises diagnostic petrochemical characterization of the lamproites based on the data reported in this paper. Until fresh lamproite samples can be analysed (conceivably from the new lamproite bodies discovered very recently in this area), any petrogenetic constraints or modelling based on the very limited data presented by Chalapathi Rao et al. (2004) can only be regarded as indicative, not definitive. Our comments on their work are, however, primarily directed towards the Nd–Sr isotope data and the interpretations thereof in terms of distinct mantle sources for the kimberlites and lamproites. We note that the Rb–Sr and Sm–Nd data reported in table 5 of Chalapathi Rao et al. (2004) are exactly the same as those reported by the same researchers 6 years previously (Chalapathi Rao et al., 1998a) and are not new data as apparently implied. The key issue is the appropriate age correction of these data, aspects of which have been previously debated by Gopalan et al. (1999) and Chalapathi Rao et al. (1998b, 1999b). Chalapathi Rao et al. (2004) provide a brief summary of the available age constraints (including references to our own studies—Anil Kumar et al., 1993, 2001) for the emplacement age of the lamproites and kimberlites (p. 913). We disagree strongly with their choice of ages ( 1400Ma and 1090Ma, respectively, for the Mahbubnagar and Anantapur kimberlites and 1418Ma for the Cuddapah Basin lamproites) used to correct the measured Nd–Sr isotope compositions to initial ratios. Conventional K–Ar ages of Precambrian rocks, even those based on multiple sample analyses, are not reliable for precise age correction of measured Sr and Nd isotope ratios. Chalapathi Rao et al. (1996), nevertheless, relied on preliminary K–Ar ages for just one sample from each of the two kimberlite fields [Kotakonda (Mahbubnagar) and Mulgiripalli (Anantapur)] and two lamproites (Chelima and Ramannapeta) to claim that the kimberlites in the Mahbubnagar cluster and all the three lamproite bodies (including the undated Zangamarajupalle lamproite) were emplaced contemporaneously at 1400Ma—more than 300Myr earlier than the Anantapur kimberlites [previously dated based on more reliable Rb–Sr phlogopite ages close to 1100Ma by Anil Kumar et al. (1993)]. Chalapathi Rao et al. (1999a) subsequently reported Ar/Ar plateau ages for groundmass phlogopites separates from the Kotakonda kimberlite and Chelima lamproite of 1401 5Ma and