No statistical study of urbanization is possible unless adequate note is taken of the census definition of ’town’ which varies from country to country and from one census year to another. In Greenland, for example, a place with 300 or more inhabitants is called an urban area while in the Republic of Korea, an urban area must have at least 40,000 inhabitants, which shows how shaky international comparisons of the level of urbanization can be in the absence of definitional adjustments. Even in the same country, there are frequent modifications of the definition of ’urban’ which call for numerous adjustments to attain comparability over time. This, for example, was the case in the U.S.A. where a new definition of ’urban’ was adopted in 1950.

Turning to India, we find that the census definition of ’town’ remained more or less the same for the period 1901-1951 and it was only in 1961 that several modifications were introduced to make the definition more satisfying from the statistical point of view. But an interesting feature of the Indian census has been the latitude given to Census Superintendents in regard to the classification of places on the border-line of ’rural’ and ’urban’. We shall deal with this aspect here and also refer briefly to the impact on urbanization of the new definition of ’town’ adopted in 1961.

The urban population comprises the total population of all the towns. The definition of ’town’ has remained the same all through the decades 1901-1951, but uniformity has not always been maintained in the application of this definition due to the inherent weaknesses in I the definition of itself and this has, to some extent, vitiated comparaability.