Despite the low variability about the general 

 mean of CPUE, we examined the data to deter- 

 mine whether the monthly means for 6 peak 

 months (October-March) for some years were 

 consistently north or south, or, east or west of 

 the overall mean. This determination was 

 made by counting for each year the number of 

 months that were to the east of the mean and 

 the number west of the mean. We used a simi- 

 lar procedure to compare, for each year, the 

 number of monthly means to the north and the 



number of monthly means to the south of the 

 general mean. These enumerations are set 

 forth in table 4. First, with respect to the east- 

 west comparisons, we note that the X^ test for 

 homogeneity was nonsignificant ( y^^ = 20.2, 

 12 d.f.). As a result the data were pooled, 

 yielding a ^^ of 0.32 (1 d.f.), which is also not 

 significant and suggests that there is no evi- 

 dence to indicate a consistent pattern for most 

 months, in any year, to be to the east or west of 

 tlie overall mean. Tor the north-south com- 



40' 

 N 

 35' 



35° 



N 



30° 



•60 58 



FEBRUARY 



155° 160° 



LONGITUDE 



Figure 1 4. --Lat i tud i nal and longitudinal moments of CPUE for each month and 

 year. The center of the cross represents the average position of the lati- 

 tudinal and longitudinal moments for each month. 



19 



