FISHERY BULLETIN: VOL. 82, NO. 2 



that reflect gear selection for large fish or incom- 

 plete recruitment of smaller fish; mean sizes in 

 those collections were as large as or larger than 

 means in subsequent collections and seem to be 

 outliers (Fig. 7A, C). Coefficients of determination 

 were higher when the earliest two collections were 

 excluded (Fig. 7A, C). 



The period of successful spawning spans 60-90 d 

 within the major September through November 

 interval. Based on mean 99% confidence limits for 

 observations and growth per day in each April- 



A. Fall78 



All Collections: 



y=- 13- 44924x- 00031 x- 

 x } -- 90 

 Two Collections Deleted: 



y = - 21+ 56983x- 00049x- 

 r'= 91 



z 



LU 



< 



o 



r- 



200 



l 80 

 160 



140 



120 

 1 00 

 80 

 60 

 40 



20 







mtial Two Collections Deleted 



95% Confidence Limits I t 

 About The Observed Mean j t 



Mean 



Range 



60 120 180 240 300 360 42U 480 540 600 660 720 



AGE (DAYS) 



June period, calculated durations of fall spawning 

 periods were 76 d in 1977, 76 in 1978, 87 in 1979, 

 and 64 in 1980 (Table 2). These estimates fall 

 within and agree with the broad 90-d duration of 

 the fall-spawning period indicated by gonad 

 maturity and weight data. 



The predominant fall-spawned groups are pro- 

 duced by fish that first spawn when 12-14 mo old 

 and the minor spring-spawned groups by fish 19-21 

 mo old. Fall-spawned fish apparently do not spawn 

 when 5-7 mo old in their first spring, because GSI 

 values for fall 1979 and fall 1980 fish remained low 

 during their initial springs (Fig. 8B, C). Peaks in 

 GSI values from September through November in 

 1979 and 1980 were formed by fall 1978 and fall 

 1979 fish, first spawning at 12-14 mo of age (Fig. 

 8 A, B). Peaks in GSI values from April through 

 June in 1980 and 1981 were formed by fall 1978 and 

 fall 1979 fish that spawned when 19-21 mo old (Fig. 

 8A, B). We were not able to determine age when 

 spring-spawned groups spawn, because these fish 

 were clearly identifiable only until 3-5 mo old (see 

 section on Age Determination and Growth). 



FIGURE 7. — Mean observed and predicted sizes at age (days) 

 for the 1978 (A), 1979 (B I, and 1980 (C ) fall-spawned groups of 

 Larimus fasciatus. Mean sizes at age were regressed on age 

 scaled to calculated hatching dates of 27 July and 15 October 

 1978, 17 September 1979, and 27 August and 17 October 1980. 

 Observed mean lengths and their confidence limits are from 

 Table 3. Regressions were significant at a = 0.001. 



ONDJ FMAMJ J ASONDJ FMAMJ J ASO 



1 — i — ' — i — i — i — i — i — i — i— i — i — i — i — i — i — i i—j I ' ' ' i__i 



1978 I 1979 I 1980 



COLLECTION DATE 



B. Fall79 

 All Collections: 



y= .13+ 53162x-00045x- 

 r J =.94 



60 120 180 240 300 360 420 480 540 600 660 720 



AGE (DAYS) 



SONDJ FMAMJJASONDJFMAMJJAS 



' — I — i — i — J — i — i — i — i — i — i i i i i i 



1979 I 1980 I 1981 



COLLECTION DATE 



(3 



Z 



LU 



< 



o 



180 

 160 - 

 140 

 120 

 100 



80 

 60 - 

 40 - 



20 

 



C. Fall80 



All Collections: 



y 09* 36704x 

 r'~- 82 

 Two Collections Deleted: 



y = - 22 + 45303x 

 r'= 93 



»/ 



/■•' 



* 



d 



i." 





-Initial Two Collections Deleted 

 ■All Collections Included 



95% Confidence Limits 1 1 

 About The Observed Mean | t 



Mean 



Range 



J_ 



_i_ 



_i_ 



-L- 



_l_ 



60 120 180 240 300 360 420 480 540 600 660 720 



AGE (DAYS) 



ONDJ FMAMJ J ASONDJF MAMJ J ASO 



i — I — i — i — i I i I I i i 'I'' i_i i 



1980 I 1981 I 1982 



COLLECTION DATE 



350 



