For the past 150 years, organic chemists were concerned with the understanding of the covalent bond. Recently, research on molecular recognition (generated by weak, non-covalent interactions) has been recognized worldwide as an important intellectual and technological frontier. Endo- (generated by convergent cavities) and exo (generated by larger bodies of similar size and shapes, or surfaces) molecular recognition, preorganization and self-organization provide the basis of spontaneous generation of functional supramolecular architectures via self-assembly from their components. It is now accepted that molecular recognition directed synthesis and self-assembly are responsible for the generation and properties of biological systems. This research aims to use molecular recognition both to self-assemble synthetic supramolecular liquid crystalline polymers and to direct their phase behavior. Two novel classes of polymers will be investigated: functional supramolecular polymers which self-assemble by using principles that resemble those of tobacco mosaic virus, and both cyclic main-chain polymers, as well as polymers containing liquid crystalline cyclophane (i.e., cyclic derivatives of main-chain liquid-crystalline oligomers) receptors as structural units. The second class of liquid-crystalline polymers will display molecular recognition directed phase transitions. It is expected that this research will produce molecular devices such as self-assembled supramolecular synthetic ion channels and various other systems which, by analogy with natural biological systems, will combine selective recognition with external regulation. Most important, this research will enable a step ahead in understanding some of the processes that nature uses and about which we know so little, and transplant them to the field of synthetic su pramolecular polymers.