Computation of urban growth for the six size categories, using their respective population from different censuses, when the towns in these categories vary widely at different points of time, has created a false alarm regarding the high growth of large towns in India. However, the quick calculations attempted by scholars claiming to correct this flaw and to establish an absence of size-growth relationship, does not seem to be free from methodological or other errors. The present paper obtains the growth rates for different size class of towns taking the towns that belonged to each class in the base year. The growth rate of class I towns works out as consistently higher than those of lower categories. It is further shown that the percentage of towns registering more than sixty per cent decadal growth is much higher in the case of the top category. The disparity in the growth rate is also low for towns of this category, which has further gone down in the seventies suggesting strengthening of their economic base. The declassification of a larger number of small towns, high variation in their growth rates, a smaller percentage of the towns registering a growth rate higher than sixty per cent, etc, on the other hand, are indications of their unstable economy. The differences in the pace and pattern of growth at the two levels of urban hierarchy seems to be a reflection, in space, of the dual structure of the Indian economy.