The Central Indian Ridge (CIR), between 18° and 20°S, shows topographic and chemical characteristics, which suggest interaction of the ridge with a mantle plume. In order to investigate the previously postulated input from the Réunion plume (presently located ∼ 1000 km off-axis to the west) on the CIR, we present chemical and isotopic compositions of basalts, collected on and off the CIR axis between 18° and 20°S.We distinguish two geographical groups of samples, called On-Axis and Gasitao, respectively. The On-Axis group is characterized by unradiogenic Sr and Pb isotope ratios and high εNd values. Gasitao group basalts have lower SiO2, are more depleted in incompatible elements and have more radiogenic Sr–Pb isotope ratios and lower εNd values than the On-Axis group. The two groups form two distinct, subparallel linear arrays in 207Pb/204Pb–206Pb/204Pb space. While the Gasitao array trends towards Réunion plume compositions, and therefore appears to contain some Réunion-type plume material, this is not the case for the On-Axis array. Along the ridge axis, Pb isotopes become more radiogenic from south to north, and incompatible trace elements become more enriched, but the compositional field of Réunion lavas is not a suitable end member for the Pb isotope and highly incompatible element trends (e.g. Ba/Nb). This indicates that the geochemical enrichment seen in the On-Axis region is not related to Réunion-type plume material.Basalts from both groups show both positive and negative Eu anomalies, which are strongly correlated with Sr/Nd ratios, thus indicating both gains and losses of feldspar phenocrysts. However, this has little effect on ratios of other trace elements. The trace element enrichment patterns are strongly correlated with Pb isotope ratios, with the most E-MORB-like samples having the most radiogenic Pb isotopic compositions. Using the trace element (TE)/Pb ratios versus 206Pb/204Pb correlations, and by extrapolating these linear correlations to TE/Pb = 0, we constrain possible 206Pb/204Pb ratios of the enriched and depleted endmembers. These lie at 18.3 ≤ 206Pb/204Pb ≤ 18.8 for the depleted and enriched components, respectively, and not very far outside the range of the actual data. We infer that the CIR MORB, between 18° and 20°S are generated by partial melting of a heterogeneous source consisting of an enriched component and a normal, depleted upper-mantle peridotite. The nature of the enriched component is a matter of speculation. As noted, its composition is different from known Réunion plume compositions. Instead, it may represent recycled (oceanic) crustal material, perhaps derived from a subducted oceanic island. It could also be formed by a “metasomatic” enrichment process similar to that modeled by Donnelly et al. [K.E. Donnelly, S.L. Goldstein, C.H. Langmuir, M. Spiegelman. Origin of enriched ocean ridge basalts and implications for mantle dynamics. Earth Planet. Sci. Lett. 226 (2004) 347–366] to explain “E-type” MORB compositions. In either case, the location of the enriched anomaly on the CIR near the intersection with the Gasitao Ridge appears to be coincidental, because the Gasitao enrichment can be traced to the Réunion plume, whereas the On-Axis group enrichment cannot. We speculate that the Réunion plume flow might be deflected towards the South by the hot upwelling E-MORB mantle, because the southernmost On-Ridge sample does fall on the Gasitao-Réunion trend.