Ample empirical evidence suggests that prenotification is an important method to increase response rates in mail surveys. However, for Web-based surveys, the results on prenotification appear to be mixed. One explanation may be the lack of awareness of the initial e-mail contacts. For instance, automatic spam filters and unidentified senders of the prenotices could lower the chance of the e-mail being noticed. To increase awareness when contacting participants for Web-based surveys, it might be more effective to send the invitations directly to the respondents' cellular phones. This method seems most promising with younger persons in Germany, because this segment has nearuniversal cellular phone coverage. To compare the effectiveness of different prenotice procedures, we experimentally varied the initial contact mode in a fully crossed two-factorial design with (1) three different prenotice conditions and (2) two invitation and reminder conditions. University freshmen were recruited for a Web-based access panel, by completing a short paper and pencil questionnaire in a classroom setting consisting of basic demographics, psychographics, and questions about lifestyles. A total of 560 students who revealed both their e-mail address and mobile phone number and agreed to participate were randomly assigned to one of the following experimental conditions: On the prenotice factor, one group received no prenotice at all, a second was contacted by a short text message (SMS) on their mobile phone, and a third group was contacted by e-mail, both announcing the prospective survey. For the survey invitation and reminder, respondents were randomly contacted either by e-mail or by mobile text messaging. Our paper will explore the extent to which these experimental manipulations influenced the sample composition, different nonresponse patterns, and the responses to the substantive questions. Practical implications for contacting participants of Web-based surveys will be discussed.