The well-known s-shaped diffusion of technology curve generally works well in developed countries. But how does it perform in the very different context of developing countries? Across a wide range of new technologies imported from the developed countries it works poorly. In most cases the penetration rate fails to reach 25% of the population. The reason for this as I see it has to do with the concentration of innovations in the rich countries and the devotion of R&D to rich rather than poor country problems. I redrew the s-shaped curve to reflect these facts. At the other extreme, however, are technologies such as the mobile phone, which have reached a penetration rate in some developing countries that is higher than in certain developed countries. The underlying reason for this unusual experience is thought to be leapfrogging, the conditions for which are especially favourable in the case of mobile phones. Therefore there is a need to redraw the curve that explains the diffusion of most new technologies in developing countries. (Contains 5 tables, 3 figures, and 3 footnotes.)