FISHERY BULLETIN: VOL. 87, NO. 1 



Table 5.— Mean fork lengths (mm) at capture and mean back-calculated fork lengths 

 at ages from whole otoliths of male king mackerel. 



Table 6.— Mean fork lengths (mm) at capture and mean back-calculated fork lengths at ages from whole otoliths 



of female king mackerel. 



Van Oosten (1929) listed assumptions involved in 

 the use of hard parts to determine age of fish: 1) 

 the structures used are constant in number and iden- 

 tity throughout the life of the fish, 2) the ratio of 

 structure size and fish size (length) remains constant 

 with growth, and 3) marks (rings) are annual and 

 form at about the same time each year. The first 

 assumption is not in doubt for otoliths. Supporting 

 the second assumption are the correlations between 

 fish length and otolith radius, which were signifi- 

 cant for whole and sectioned otoliths but stronger 

 for the latter. It is in meeting the final assumption 

 that the validity of ages from whole otoliths becomes 

 doubtful. The distributions of focus-ring measure- 

 ments were only slightly better for sectioned than 



for whole otoliths. However, the distribution of 

 monthly percentages of whole otoliths with opaque 

 margins was multimodal, indicating nonannual ring 

 formation (or large and numerous reading errors), 

 while that of sections was unimodal and fairly nor- 

 mal, indicating annual ring formation peaking in 

 August-September. Manooch et al. (1987) found a 

 peak in ring formation during February-May, but 

 they also found ring formation in September for 

 some fish taken off northwest Florida and suggested 

 that this difference may be due to separate spawn- 

 ing groups within the Gulf of Mexico. 



We consider rings in otolith sections valid annuli, 

 but our evidence for validation is indirect, as in 

 previous studies of king mackerel. As pointed out 



58 



