The original 1950 estimate for the trailer-type boats was based on 

 the 1888 boat permits issued by the National Park SeiTTice. From obser^ 

 vations and contacts with this type of boat fisherman^, it was presumed 

 that each boat took at least one limit of fish per day's fishing* As 

 some boats are used on the lake for only one or two days and others for 

 two weeks or longer^ it was further presumed that the average number of 

 days of fishing on the lake was three. 



Bealizing the limitations of the 1950 estimate, a more extensive 

 survey was made in 195le For collection of more accurate data^, a vol-= 

 unteer creel census for trailer-boat anglers was set up in 1951 • A 

 card with questions pertinent to the study was attached to each boat 

 permit issued. The cards were to be turned in. at the gate as the party 

 left the Park, Of the 2056 boat permits and cards issued^ the returns 

 totaled 6l7j, or 30 percent* 



At the same time, and as a check of the validity of the volunteer 

 card returns, road blocks were set up one day each week^ and all cars 

 with boats were stopped. The road blocks were so placed that only cars 

 leaving the Park were stopped; thus, only these boat fishermen who had 

 completed their fishing activities were interviewed. The same and ad= 

 ditional information requested on the volunteer cards was gathered con= 

 cerning the use of the boat on Yellowstone Lake, Sunday was generally 

 chosen as the day for the road block because many of the boats on 

 trailers are brought into the Park from nearby areas for weekend fishing 

 only. 



The season of 1951 was divided into four periods of approximately 

 one month each, and the results of the volunteer census and the road 

 block were compared. For each period the means were calculated for the 

 number of days of fishing on the lake, the number of persons fishing 

 from the boats the number of hoiurs the boat was used each fishing day^ 

 and the catch-per-hour of fishing. The application of the "t" test to 

 the l6 possible comparisons indicated no significant difference in 13 

 cases (table lii). Two of the three differences were in the number of 

 hours the boat was used each fishing day, and the third difference was 

 in the number of persons fishing from the boat. As the majority of the 

 values of "t" were not significant, it was concluded that the results of 

 the two methods be considered as the same. Because the amount of data 

 from volunteer census cards was considerably greater than that of the 

 road block interviews, the results of the cards (table 15) were used in 

 computing the trailer boat estimates for the 1951 season. 



