Long before nanotechnology, the semiconductor industry was miniaturizing microelectronic components. Since the late 1950s, that industry's dominant material has been silicon. Yet there have always been competitors to silicon that supporters hope will upend the semiconductor industry. It is impossible to understand this industry without a more complete picture of these alternatives — how they come about, how they capture organizational support, why they fail. It is equally impossible to understand nanotechnology without a focus on these alternatives, since research communities devoted to perfecting them today form the backbone of the nanotechnology field. We trace the history of the longest lived silicon alternative — molecular electronics. Molecular electronics arose in the late 1950s as a visionary program conducted by westinghouse on behalf of the air force. We attribute its failure to the difficulties inherent in matching a futuristic vision to a bureaucratically accountable, incremental program that could compete with silicon. Molecular electronics reappeared again at ibm in the 1970s and at the naval research laboratory in the 1980s. In each of these incarnations, molecular electronics' charismatic champions failed to gain the organizational support to make it a mainstream technology. Only at the turn of the century, with new nanotechnology institutions and new models of industry—university collaboration, has some form of molecular electronics neared acceptance by the semiconductor industry.