In this paper we investigate the potential production and implications of a global biofuels industry. We develop alternative approaches to the introduction of land as an economic factor input, in value and physical terms, into a computable general equilibrium framework. Both approach allows us to parameterize biomass production in a manner consistent with agro-engineering information on yields and a "second generation" cellulosic biomass conversion technology. We explicitly model land conversion from natural areas to agricultural use in two different ways: in one approach we introduce a land supply elasticity based on observed land supply responses and in the other we consider only the direct cost of conversion. We estimate biofuels production at the end of the century will reach 220 to 270 exajoules in a reference scenario and 320 to 370 exajoules under a global effort to mitigate greenhouse gas emissions. The version with the land supply elasticity allows much less conversion of land from natural areas, forcing intensification of production, especially on pasture and grazing land, whereas the pure conversion cost model leads to significant deforestation. The observed land conversion response we estimate may be a short-term response that does not fully reflect the effect of long-run pressure to convert land if rent differentials are sustained over 100 years. These different approaches emphasize the importance of reflecting the non-market value of land more fully in the modeling of the conversion decision.