WE are happy to see a critical scrutiny of our paper [Kakwani and Subbarao 1990] by S Tendulkar and LR Jain (TJ for short hereafter)[Tendulkar and Jain 1990]. The measurement of poverty involves a number of conceptual and practical difficulties. Many a time we need to settle for second best methods because of non-availability of appropriate information. For instance, like most other researchers on Indian poverty, we have used per capita household expenditure (PCHE) as a measure of household economic welfare. A better measure of household welfare will, of course, be the per adult equivalent consumption which corrects' for the differing needs. of adults and children. But this measure cannot be employed in Indian studies because the NSS data are available to researchers only in grouped form (the groups formed on the basis of per capita household expenditure), although the NSS organisation collects the expenditure data for each household. The grouping involves considerable loss of information which may lead to biased estimates of poverty. To estimate poverty from such data, one needs to employ some interpolation device. Most Indian studies have employed a two-parameter lognormal distribution [Minhas, Jain, Kansal and Saluja 1987], with the exception of Ahluwana (1978) who employs Kakwani-Podder's [1976] Lorenz function. Since the NSS does not regularly correct the income ranges in order to take account of inflation, inappropri, ate interpolation devices may induce large estimation errors. These errors will be particularly serious when one uses a single density function such as lognormal to the entire consumption range. In our study we used a general interpolation device [Kakwani 1980] which uses a different density func-tion within each consumption range. Although this procedure is an improvement over those employed by previous researchers, it is still the second best solution. TJ do not seem to recognise these and many other problems associated with pover-ty research in India. Had they appreciated these, they would have been more constructive in their evaluation of our paper; instead they adopted the negative approach of attacking-wrongly and unfairly in most instances as we shall soon demonstrate-everything in the paper. In what follows, we respond to their criticisms not in the order chosen by them, but in order of the impor-tance of the issues raised by them. These fall into five groups:(a) our choice of price deflators;(b) problems with the decomposi-tion methodology;(c) growth elasticities;(d) regression results; and (e) other miscellaneous issues including validity of our conclusions.