The microstructures of highly isotactic polypropylenes, prepared with magnesium chloride. supported Ziegler-Natta catalyst systems, have been determined experimentally through C-13 NMR measurements and also predicted through Markovian statistical models. Polypropylene molecules are seldom 100% isotactic but possess long isotactic sequences interrupted by stereo- and regio-irregularities called chain defects that ultimately define the crystallinity of the polypropylene. The structures and numbers per 10000 repeat units of the various interrupting stereo-irregularities, termed "stereo-defects", are determined in this study. It is shown that different families of stereo-defects in highly isotactic polypropylenes can be associated independently with symmetric and asymmetric chains. The single-parameter-zero order Markovian statistical approach typically utilized in the Doi two-state asymmetric and symmetric chain statistical models for characterizing polypropylene sequence distributions has been extended to first-order Markov for symmetric chains. First-order Markovian statistics for asymmetric chains naturally reduce to zero order for the isotactic polypropylenes examined in this study, whereas the symmetric chain components do not. Syndiotactic blocks of steadily increasing sequence lengths are predicted by first-order Markovian statistics for highly isotactic symmetric chains possessing a meso diad content of 0.99. The average sequence lengths of these syndiotactic blocks are predicted to become shorter as the polypropylenes become more isotactic. Finally, it was observed that the sequence distributions favored asymmetric chains with the addition of increasing amounts of an electron donor to the supported Ziegler-Natta catalyst system.