This theoretical study was inspired by the perpetual debate over the so-called ''dormancy'' of the active sites in propylene polymerization, i.e., a drop in their activity after a regioerror (2,1-insertion), which was reported to occur in many (although not all) catalytic systems. To explore the range of possible situations, we have selected two homogeneous systems of fundamentally different structure: an octahedral system of C2 symmetry with a tetradentate OOONONOOO ligand and a bridged indenyl catalyst. This choice was not accidental; it is in these two systems where the experimentalists cannot reach a consensus about dormancy. Our densityfunctional theory calculations explain why in certain catalytic systems both primary and secondary alkyl complexes can be equally reactive toward propylene polymerization, despite the intuitive concept of dormancy. To understand such a behavior, it was imperative to build an extensive model, including the counteranion and solvent effects. The discussion is also supplemented by our latest calculations on the classical second-generation Ziegler-Natta system. catalyst ͉ density-functional theory ͉ olefin polymerization ͉ propylene ͉ transition metals T he idea of ''dormancy'' of the active sites in polymerization of propylene appears to be straightforward and selfexplanatory. Propylene can insert into the metal-carbon bond in two modes, namely 1,2-insertion and 2,1-insertion (see Fig. 1). These modes lead to the polymer chains with the primary and secondary carbon atom attached to the metal center, respectively. It seems obvious that the secondary C ␣ atom creates some steric congestion around the transition metal atom, thus making further insertion events less favorable. As a result, 1,2-insertion becomes predominant, and the concentration of secondary alkyl complexes increases because they are consumed in the process of polymerization at a relatively low rate. Yet the experimental results are not so consonant.Recently, it has been demonstrated that in the regioselective catalysts the occasional regioerror decreases the rate of a subsequent propylene insertion event by a factor of 100-1,000 (1); however, there is no difference in reactivity with respect to ethylene. This series of study, initially performed for two siliconbridged indenyl systems (2), has been extended to include the predecessor of heterogeneous catalysts (1) and two modern systems with a Lewis base (3). The plentitude and miscellany of systems displaying dormancy, discussed in refs. 1-3, suggest that this phenomenon is probably common, if not ubiquitous.However, the antithesis based on a carefully designed experiment, which demonstrated that the activities of primary and secondary alkyls are comparable for the carbon-bridged indenyl catalyst, was formulated almost immediately (4). The follow-up study (5), showing one more example of potentially dormant postmetallocene catalyst, originally developed by Kol et al. (6), is probably not the last experimental paper dealing with the subject.The polemic described abov...