One of the first major efforts to develop ways of obtaining wilderness use informa- 

 tion was made by Wenger and Gregersen in the Three Sisters and Mountain Lakes Wilder- 

 nesses in Oregon during 1961 and 1962. Wenger (1964) tested the general effectiveness 

 of unmanned, self-registration stations by placing two different types of registration 

 boxes and forms and three different types of signs on trails and then interviewing 

 visitors up the trails, out of sight of the registration stations. He concluded that 

 self-registration was a promising approach to wilderness use estimations; that a single 

 card per party, dropped into a box after completion, was superior to a book-type 

 registration system; and that a firm but explanatory sign was best. Registration rates 

 averaged 74 percent for both years and both areas. Location of registration stations 

 had an important influence on response rates. 



Wenger and Gregersen (1964) also reported on the association of registration rates 

 and various visitor and location characteristics. They concluded that "raw" registra- 

 tion data contained biases, specifically underrepresenting riders, fishing and elk- 

 hunting parties, one-person parties, and repeat visitors. Many other factors--including 

 trails, day of week, size of residence community and distance to it, length of stay, and 

 occupation- -were unrelated to registration rates. They also concluded that the effects 

 of nonresponse must be fully considered, and that modifications must be made in findings 

 stemming from raw registration data. Wenger and Gregersen did not report on the conver- 

 sion of registration data into use estimates. 



Lucas (1964) studied recreational use of the Quetico-Superior area in Minnesota 

 and Ontario, Canada. Traffic counters were placed on six major access roads and drivers 

 were briefly interviewed as they left the area. The sample included about 4,000 groups-- 

 approximately 4 percent of the total traffic. Estimates of recreational characteristics 

 for the total measured traffic for each road were made on the basis of the composition 

 of the sample interviews. Error terms were not calculated. Visits were shorter than 

 expected; 41 percent of the visitors did not stay overnight, and the average visit was 

 only 1.75 days. Use was also very unevenly distributed- - 10 percent of the access points 

 accounted for over half of the total visitor use for every type of user. 



3 



