This paper deals with the distribution of innovating activities across industries and technological areas in the context of a Knowledge-based Economy. The results are directly based on data concerning French firm patenting in the US (over the period 1985-1990). Each patent is identified by the industrial sector in which it is produced and by the technological field in which the patent application has been made. Innovative activities are concentrated in a few industrial sectors and in a few technological fields. Our main contribution concerns techno-industrial clusters (blocks of technological knowledge structured by a set of industrial sectors). On the basis of an algorithm (Zig-Zag) we identify several techno-industrial clusters, three of which underpin the French system of innovation : chemicals, electronic and communication, transport and mechanical engineering. Our analysis shows that the strong patterns derived from the most complex clusters deal with sectors that invest intensively in R-D. Such a result confirms the view concerning the emergence of pervasive technological families .