i the information on the assumptions made on the average thickness and the average grade considered in the estimate. It is stated that 'the reefs show strike persistence from 10 metres to a couple of hundred metres'. If so, how the three reefs with 3.7 gm, 5.0 gm and 4.0 gm with thickness of 20 cm, 30 crn and 35 cm respectively could be expected to extend for 2.5 km. Though, the authors have mentioned about the persistence PGE mineralization in 8 boreholes, only 7 boreholes have been shown in the map (Fig.2) over a strike length of <2 km. The authors have not provided information on the grades and thickness of mineralization in any other boreholes. From the map and the description, it is inferred that there is only one trench closer to the BH1 profile in the entire area and there are no other surface controls of mineralization. Though, there are indications of a number of zones of PGE mineralization in the trench, it is not known which one of them is correlatable with the reefs intersected in the BH1. It would be beneficiaI to the readers, if these information are also provided. In the absence of the same, it can only be presumed that the grades and thickness of the reef intersected in BH-1 has been extended for the entire length and the figure of 7500 kg of combined Pt+Pd may be considered only as resource potential rather than 'reserve'. It has been stated that 'only the central fine-grained ultramafite and the chrorniferous lenses in the eastern magnetite seem to host PGE mineralization of commercial interest with Pt+Pd values ranging from 1.5 to 6 ppm' (p.535, abstract). However, it may be noted that the richest mineralization is not on the eastern side but on the foot wall (western side) of the magnetite band (Fig.5). The finegrained ultramafite (chromite-chlorite schist) is considered as meta-dunite with variable proportions of chromite, while the coarse-grained ultramafite as equivalent of metapyroxenite with variable proportions of Fe-Ti oxides. Extreme variations of the rock corresponding to chromitite and chlorite schist have been reported in the fine-grained ultramafite. The chromi tite bands are also shown in borehole log of BH1 (Fig.5). The scale used for plotting the Cr values is logarithmic. It starts with 0.01 followed by 0.1 and the next value should be 1 % instead of 10%. Even if it is correct, it is surprising that none of the samples analyse more than 10% chromium. The single grain data of chromite (JS-159; Table 6) also shows a Cr,03 value of about 30% and Fe,O, varying from 19.44 to 34.56% (total FeO ranges from 49.73 to 61.87%) with CrIFe ratio of 0.43 to 0.55. The Cr,O, content of all the fine-grained ultramafites are less than 17.03% (Table 7). More than 70 bore hole and surface samples analysed by XRF method have indicated values in the range of 0.03 to 0.06% except for one sample with 9.44%. The EDS and EPMA studtes carried out by us have shown that the oxide grains are found to be magnetite and chromiferous magnetites and there is no chromite sensu stricto. Thin section samples of the ultramafic units of the complex did not reveal presence of any relict olivine. Presence of chlorite do not confirm that the parent rock was dunite. Moreover, no serpentinisation is seen which is common in other ultramafic complexes. The EPMA studies have confirmed relict pyroxene (hypersthene, diopside and augi te). So, the host for the fine-grained ultramafi te can also be a pyroxenite as in the case of coarse grained ultramafi te. The authors have attempted a correlation of the PGE mineralization of Hanumalapur segment with that of the UG2 chromitite of Bushveld Complex. It is like comparing mole hill with mountain. The UG2 chromitite layer with 0.5 to >1.0 m thickness and extending for hundreds of kilometers has a chrome content of 43.5% Cr,O, and a CrFe ratio of 1.26 to 1.4 (Schurmann et al. 1998).