Period 3: 

 (September- 

 December) 



These months of recruitment 

 through molting show increasing 

 catch-per-unit-of-effort values 

 on a monthly basis with pro- 

 gressively lower ocean tempera- 

 tures. In this case we conclude 

 that recruitment through molt- 

 ing with the hypothesized vul- 

 nerability has a far greater 

 effect than water temperature. 

 While this is plausible for the 

 months of peak molting (August 

 and September in 1968; see 

 analysis of cluster samples), it is 

 difficult to accept a continuing 

 high recruitment and vulnera- 

 bility during October and 

 November when the molting 

 percentages decrease. 



Continuing with the analysis, we next used 

 catch -in -numbers -per -trap -haul -set-over -day 



with the same temperature data (Fig. 10) to 

 determine if similar conclusions would result: 



Period 1: 

 (January- 

 April) 



Period 2: 

 (May-July) 



These monthly catch-per-unit- 

 of-effort values increase along 

 with the ocean temperatures. 

 This situation does agree with 

 our premise that availability 

 should be increasing. 

 This downward concave curve 

 possibly indicates that even 

 though the monthly ocean tem- 

 peratures continue to rise, the 

 year- or molt-class strength from 

 the preceding year diminishes. 

 Incidentally, there is not as wide 

 a disparity between April and 

 May in catch in numbers per 

 trap-haul-set-over-day as there 

 is with catch in numbers per 

 trap-haul. 



1968 



< 



> 



O 



3 

 < 



(PERIOD 3| 



2 



z 

 z 



I 



i— 

 < 



.2- 



IPERIOD 1| 

 'Jon 



IPERlOD 2| 



< 



s 

 s 



30 32 34 36 38 40 42 44 46 48 50 52 54 56 58 60 62 

 SURFACE WATER TEMPER ATURE I °f I 



Figure 10. — The relationship between mean ocean surface temperature and 

 catch in numbers per trap-haul-set -over-day by month for 1968. 



34 



