REVIEW ARTICLE 97 skewed than the distribution of owned holdings. At the outset, we must point to the incomparability of the two distributions, since operated holdings are inclusive of the land leased out by absentee landowners, whereas these are excluded from the ownership distribution. Hence, the size class of the absentees (who own 20% of operated holdings) is crucial in validating the result. Regulated tenancy, an outcome of a strong tenants' movement, is offered as the explanation for this observation. However, in the absence of information regarding the size-class of registered tenancy, it is not.irrefutable that such tenants are indeed small/middle peasents. Given the evidence of such operated holdings being used as collateral and once and for all fixed cash rents, a case can also be made for including these holdings in owned area, or at least being excluded from leased area. The second observation which merits explanation is the remarkable equality in level of regulated and unregulated rents which are not 'high*. This cannot be attributed completely to a strong tenants' movement, as argued by the authors, but perhaps confirms the earlier inference of a low demand for commercial leasing. What is more puzzling is the stickiness of open market cash rents over time. Finally, a tendency towards deconcentration is discerned in comparisons between inherited and presently owned land, especially marked in the wet ecotype. While this may be a once and for all outcome of sales to circumvent land reforms legislations, it does not indicate that the oft repeated argument which views dispossessions— of ownership and access to operated holdings—as concomitant with 'capitalist* or commercialised production may be incorrect. This conclusion raises important methodological issues in defining capitalism and its traverce—that capitalist development may take contextual forms distinct from certain standard Eurocentric models. Labour relations are viewed in terms of the prevalent forms of hired labour and changes therein, income and employment, non-agricultural work, poverty and a concluding section comparing hired and family labour. They argue that 'wage labour relations in this economy are quite heterogenous, and they can hardly be described with the single dichotomies of permanent-casual and skilled-unskilled' (p. 148). Though they do not go beyond arguing for various forms of semi-attachment between attached and casual workers, they can in fact take the discussion many steps forward by demonstrating the impossibility of identifying a specially capitalist form of hired labour. This argument is made tenable by their documentation of the emergence of the 'exclusionary labour arrangement', of Kothu gang labour in the wet area which act as proto-unions, and has spread at the expense of daily wage labour and reduced the bargaining power of permanent farm servants. Interestingly, permanent farm servants, some of whom are