GOOD DEAL of scholarly attention has been paid in recent years to the increasing disaffection of the American electorate. Much of this concern has centered on the sources and meaning of political discontent, such as the erosion of political trust. Some have interpreted this trend as transient dissatisfaction (Citrin 1974); others have been considerably less sanguine (Miller 1974a, 1974b). In any case, political trust, taken by itself, is not a strong predictor of levels of political participation (Muller 1977; Milbrath and Goel 1977: 68-74). Better behavioral explanations are obtained when political efficacy also is taken into account (Aberbach 1969; Miller and Miller 1975: 406-7). It appears that, when low levels of trust are combined with high levels of efficacy, the potential for unconventional or "nonallegiant" action is optimal. Presumably, efficacy affects the level of initiative required by the act, and trust structures its allegiant or nonallegiant nature. This is the familiar "mistrust-sense of political efficacy" hypothesis, and it has been widely investigated (Abravenal and Busch 1975; Craig 1979; Craig and Maggiotto 1981; Finifter 1970; Fraser 1970; Hart 1978: chap. 2; Hawkins, Marando, and Taylor 1971; Paige 1971; Watts 1973; Zurcher and Monts 1972). Of course, in his original formulation of the hypothesis, Gamson (1968) connected this cynical combination of beliefs to all forms of activism conventional and unconventional. Even so, most studies have shown low trust and high efficacy to be related to nonallegiant action, but not to conventional participation. However, the development of cumulative research in this area has proven difficult for several reasons. One problem was mentioned above. We know that political trust is declining, but there is still some question whether we are witnessing a fundamental rejection of political institutions or a less durable disapproval of government officials and policies (Abramson and Finifter 1981). Clearly, our behavioral predictions depend on this distinction: Conditions of low diffuse support will foster different participatory consequences than conditions of low specific support for the system (Craig 1980). Additionally, most studies have not been sensitive to the emergent dimensional properties of political efficacy. All evidence now strongly indicates that political efficacy itself measures an "external," system-regarding attitude, as well as an "internal," personal dimension of belief (Converse 1972; Balch 1974; Craig 1979). Finally, we lack evidence for the connections between these political attitudes and a wider range of conventional or "within-system" behaviors, other than voting (Verba and Nie 1972).