lake trout were established in M -2 in 1944, in 

 M-3 in 1945, and in M-4 in 1946. The decline 

 of the lake trout took place later in the more 

 southerly districts . Years of above -average 

 abundance occurred in M-5 and M-6 as recently 

 as 1945 and in M-7 and M-8 as late as 1947. The 

 decline of lake trout first carried the stock to 

 new record lows as follows: M-5, 1948; M-6, 

 1947; M-7 and M-8, 1949. The combined effects 

 of the introduction of nylon netting and reduced 

 predation by lake trout created a trend toward 

 higher catches per unit effort that makes unde- 

 sirable the inclusion of the more recent years 

 in the study of correlations of fluctuations of 

 availability in the several districts . 



Conditions were more stable over the 

 15 -year base period, 1929-1943. To be sure, 

 the abundance of the predatory lake trout varied 

 but the changes were not extreme (Hile, Esch- 

 meyer, and Lunger, 1951a). Furthermore, 

 changes were made in fishing laws and in the 

 kind of twine in the nets, but the shift from 

 linen to cotton did not bring increase of efficiency 

 comparable to that which .resulted from the intro- 

 duction of nylon and, as was explained in the 

 introduction to this section, the changes of 

 regulations may have had no major effect on 

 fishing success. Even though limitations of the 

 data must be recognized the abundance indices 

 of 1929-1943 offer our best materials for in- 

 quiry into the extent of correlations between 

 fluctuations in the abundance of chubs in differ - 

 ent districts. 



The correlation coefficients (table 11) 

 give some evidence of similarities between dis- 

 tricts but do not indicate lake -wide correlation. 

 Closest similarities were among the 4 northern - 

 most districts, M-2 through M-5. All 6 coef- 

 ficients were positive; 1 was significant at the 1- 

 percent level (M-3, M-5), 1 at the 5-percent 

 level (M-4, M-5), and 2 at the 10-percent level 

 (M-2, M-3; M-2, M-5). Only the correlations 

 of M-4 with M-2 and M-3 fell short of the 10- 

 percent probability. Farther south, the fluctua- 

 tions in abundance of chubs in M-6 gave no 

 evidence of correlation with those in any other 

 district. Fluctuations in M-7 were correlated 

 significantly with those in M-2(p<0.05) and the 

 correlation of M-7 with M-8 fefl just short of the 

 5 -percent level. Aside from this correlation 



with M-7, the fluctuation in M-8 appeared not 

 to be correlated with those in other areas . 



Discussion of the possible significance 

 of the correlations that were found between 

 fluctuations of chubs in different districts would 

 be little to the point, since there is no means of 

 judging the extent to which they were influenced 

 by changes in the law and in the kind of twine 

 fished and since the fluctuations involved a group 

 of species . 



Correlations were calculated also between 

 the 1929-1943 abundance of chubs of commercial 

 size and of the predatory lake trout and the re- 

 lated whitefish and lake herring (table 12). Al- 

 though the predation of lake trout on chubs bore 

 heaviest on small fish, it was not necessarily 

 limited entirely to them . One might anticipate 

 a possible negative correlation between the 

 abundance of the two . The data do not, however, 

 meet this expectation for only 3 of the 7 co- 

 efficients were negative and these had small 

 values. The 2 largest values were both positive; 

 that for M-6 was significant at the 10 -percent 

 level and the one for M-8 at the 1 -percent level. 

 Apparently the fluctuations in the abundance of 

 lake trout in 1929-4943 were not great enough to 

 exert a detectable effect on the abundance of 

 commercially valuable chubs 12/ . Some evidence 

 exists for correlation between fluctuations in the 

 abundance of chubs and whitefish . Six of the 7 

 coefficients were positive, 1 was significant at 

 the 1 -percent level (M-4), and 1 was near the 1- 

 percent value (M-7). The only indication of corre- 

 lation between the abundance of chubs and lake 

 herring occurred in M-8 where the coefficient 

 was significant at the 5-percent level (the fluctua- 

 tions of lake herring were for districts M-6, M-7, 

 and M-8 combined- -see footnote to table 12). 



12/ Further evidence that both lake trout and 

 commercially exploitable chubs can be plentiful 

 at the same time comes from data for M-8 in 

 1944 and 1945. In those years the abundance 

 indices for lake trout were 174 and 166, respec- 

 tively, and those for chubs were 299 and 205 . 



30 



