KISHKKY BL'LLETIN: VOL. 81. NO. 4 



FIGURE 3.— Slopes ±95% confidence intervals for body length (cm) — scale size regression equations of weakfish, 



Cynoscion regalis, for each area. 



significantly alter the regression. Area 5 and Area 6 

 were each significantly different (P < 0.001) from 

 other equations and were thus considered separate 

 regions. The three regional equations determined 

 from this analysis were 



Region I In TL = 1.948 + 1.139 In SS 



(areas 1-4) r = 0.883 n = 850 



Region II In TL = -1.179 + 0.974 In SS 

 (area 5) r 2 = 0.835 n = 190 



Region III In TL = -0.329 + 0.78b In SS 

 (area 6) r = 0.639 n = 200. 



Maximum age differed substantially between 

 regions but not between sexes. The greatest ages 

 occurred in Region I where both sexes attained age 

 11. Females in Region II did not exceed age 6 while 

 maximum age for males was age 5. In samples from 

 Region III, the oldest age for both sexes was 4 



yr. 



Using the above regional equations we back- 

 calculated lengths at age and tabulated mean lengths 

 by sex, age group, and region (Table 1). Mean lengths 

 for successive age groups were examined for 

 evidence of differential survivorship, i.e., Lee's 

 phenomena, using analysis of variance. Although a 

 few older groups had larger size at age 1 , there was no 

 significant increase in size of the first year (P < 

 0.05). 



An analysis of variance (ANOVA) of back- 

 calculated lengths at age revealed significant dif- 



ferences between regions for ages 1-4 (P< 0.05). The 

 regions responsible for the differences were deter- 

 mined by Duncan's new multiple range test (Table 2). 

 Region III fish were significantly larger (P< 0.001) at 

 age 1, 22 cm, compared with 20 cm for Regions I and 

 II. By age 2, the mean length for Region I was largest, 

 32 cm, and significantly different from Regions II and 

 HI at 29 and 28 cm, respectively. The difference be- 

 tween Regions II and III was not significant. Greatest 

 size differences existed for mean calculated lengths 

 at age 3, between Regions I and III at 47 and 35 cm, 

 respectively. Calculated mean length at age 4 was 

 significantly greater in Region II than Region I (P < 

 0.05), while lengths at ages 5 and 6 were not 

 significantly different between regions. 



Lengths at age were greater for females in all 

 regions and the differences increased with age. 

 These differences did not become statistically 

 significant until age 6, except for 2-yr-olds in Region 

 III where females were 3 cm larger (P< 0.05). Region 

 I had significant differences between sexes (P < 

 0.05) for ages 6-11, except at age 9. Lengths at age 6 

 were 68 cm for females {n = 53) and 66 cm for males 

 (n — 39); the comparison at age 11 was 81 cm for 

 females (n = 3) and 70 cm for males (n = 1). No 

 significant interactions were found between sexes 

 and regions. Based on this analysis, length at age data 

 were pooled for sexes in Regions II and III but 

 separated for Region I in the subsequent fitting of 

 growth curves. 



In all three regions, the greatest growth in length 



806 



