It has become a well-documented finding that social ties are a primary and possibly even an indispensable avenue of recruiting for most religious sects. What has not been clear in the literature is the nature of these social ties and whether they differ for various kinds of groups as well as individuals within those groups. An assumption in the Lofland-Stark model (1965) and in Snow, et al.'s, work (Snow, Zurcher and Ekland-Olson, 1980) is that these social networks are characterized by effective relationships of some duration. Our data show that for some groups close friendship ties are a means of recruiting while for others an acquaintance-type encounter is a major recruiting device. Moreover, our data suggest that these differences are accounted for, in part, by differences in religious beliefs among religious groups.