This paper explores the dynamics of smallholder technology adoption, with particular reference to a high-yielding, low-external input rice production method in Madagascar. We present a simple model of technology adoption by farm households in an environment of incomplete financial and land markets. We then use a probit model and a symmetrically trimmed least squares estimation of a dynamic Tobit model to analyze the decisions to adopt, expand and disadopt the method. We find that seasonal liquidity constraints discourage adoption by poorer farmers. Learning effects-both from extension agents and from other farmers-exert significant influence over adoption decisions.