316 FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE 



The fact that in all districts but H-2 the deep trap net was not fished through- 

 out the entire "period of reference" (1929-1934) introduced certain difficulties into the 

 estimation of abundance. For example, the deep trap net was operated in H-l during 

 only 5 years (1930-1934) of this 6-year period. The average catch of whitefish per 

 lift of deep trap nets in the years, 1930-1934, was 111.08 pounds. However, the data 

 for large-mesh gill nets and pound nets indicated that the 1930-1934 abundance aver- 

 aged only 99.12 percent of the 1929-1934 mean. Consequently, the average catch per 

 lift of deep trap nets would have been higher had the gear been fished in 1929 also. It 

 was necessary, therefore, to base the computations of the expected catch (p. 314) of deep 

 trap nets on the "corrected" catch per lift, 111.08/0.9912=112.07 pounds. 



Although this method of "correcting" the average catch per lift of deep trap nets 

 (in some districts the data for pound nets had to be treated similarly) is sound logically, 

 the actual reliability of the results is open to question in some districts in which the rise 

 of the deep-trap-net fishery was accompanied by the practical extinction of the gill-net 

 and pound-net fisheries (for whitefish). The correction was based, for example, on the 

 data for only 3 years in H-3 and H-5 and for 2 years in H-6. The difficulties involved 

 in following annual changes in abundance in areas in which the deep trap net replaced 

 other types of gears completely or nearly completely will be mentioned again on page 328. 



The deep trap net was important also in Green Bay and northern Lake Michigan, 

 but the disturbance of the fishery was not as severe as in Lake Huron. 



Although the greatest need for dependable statistical data existed with respect to 

 those districts in which the deep trap net became almost the only gear that produced 

 whitefish, it was for precisely these areas that the original data were least trustworthy. 

 This lack of dependability had its origin in the extensive inaccuracies and misstatements 

 of fact known to have occurred in the reports of numerous deep-trap-net fishermen. 

 This observation is not intended as an indictment of any fisherman or group of fisher- 

 men. Nevertheless, the fact that these inaccuracies existed cannot well be ignored. 

 To discuss changes in abundance computed from deep-trap-net data without giving some 

 idea as to their degree of dependability would be misleading. Misstatements were found 

 in the reports of deep-trap-net fishermen as to the type of gear fished, the numbers 

 of nets lifted, and the size of the catch. 



Numerous deep-trap-net reports were indicated erroneously to be reports of pound- 

 net operations. Most of the errors of this type were made by operators in the Saginaw 

 Bay region in 1931 and in both the Saginaw Bay and Harbor Beach regions in 1932. In 

 other years and in other districts the designation of deep trap nets as pound nets was 

 much less frequent. Without naming sources of information or explaining the pro- 

 cedure followed, it may be stated that we are certain that we have detected and cor- 

 rected practically all, if not all, of the misstatements as to the type of gear. Conse- 

 quently, this originally serious source of error does not affect materially the data of this 

 paper. 



It has not been possible to correct the inaccuracies of data as to the number of nets 

 lifted and the size of the catch, nor is there any basis for a good estimate of the extent 

 of these inaccuracies. Where there was opportunity of comparing actual and reported 

 data the discrepancies were sometimes appalling. Some fishermen not only reported 

 incorrectly the number of nets lifted but gave dates of lifting that did not coincide with 

 the dates on which they actually left port. The reported catches were often understate- 

 ments. The extreme in this type of misrepresentation is offered by the report of an 

 operator who is known to have taken more fish in a single day than he reported for the 

 entire month. It must be considered highly probable that the actual total production of 

 whitefish in deep trap nets was far above that recorded in this study. 



In calling attention to the defects in the deep-trap-net data it is not intended to 

 imply that all operators of deep trap nets submitted erroneous and carelessly prepared 

 reports. There is good evidence that many of them prepared scrupulously accurate ac- 

 counts of operation and of catch. Although the number of inaccurate reports may be 

 sufficient to invalidate the deep-trap-net data as descriptive of details, these data still 

 serve satisfactorily to indicate the trends of the fisheries in the different districts. This 

 view finds support in the fact that for the whitefish as well as for other species there 



