FISHERY BULLETIN: VOL. 75, NO. 2 



this method, because zones on the outer edge are 

 narrow and therefore difficult to distinguish until 

 late in the growing season. Moreover, because of 

 the irregular growth of otoliths of older rockfish, 

 different marginal areas provide different results. 



Otolith Sections 



Results indicate that consistency of otolith 

 readings is superior to that of scales or other 

 structures for the three species of rockfishes 

 studied, but agreement of otolith readings still 

 may be unsatisfactory. Otoliths were sectioned to 

 try to improve consistency of readings. Blacker 

 ( 1974) noted that annuli are laid down only on the 

 proximal (internal) surface of the otolith during 

 later years in the life of fishes such as sole, Solea 

 solea; plaice, Pleuronectes platessa; turbot, Scoph- 

 thalmus maximus; redfish, Sebastes sp.; and horse 

 mackerel, Trachurus trachurus. These annuli are 

 not seen when the distal surface of the otolith is 

 used for age determination and the investigator 

 underestimates the age of the fish. 



Exact agreement between readings of whole 

 and sectioned otoliths of canary rockfish (37 vs. 

 219c ) differed by 16 percentage points (Table 4). A 

 chi-square test for paired data corrected for 

 continuity revealed that there was a significant 

 difference between the two (P<0.025). Percent 

 agreement between first readings of whole and 

 sectioned otoliths was low with a value of 51% ±1 

 assumed annulus. The similarity of the mean 

 estimated ages indicates that the phenomenon 

 reported by Blacker (1974) probably does not 

 occur in canary rockfish otoliths. Ages were not 

 substantially underestimated by reading the 

 distal surface of the whole otolith. 



Sectioning did not improve consistency of 

 readings of canary rockfish otoliths. Moreover, it 

 is not possible to follow specific annuli completely 

 around the sectioned otolith to determine if an 

 assumed annulus is split. Whole otoliths allow the 



TABLE 4. — Percent agreement between first and second read- 

 ings of whole otoliths and between first and second readings 

 of sectioned otoliths, and percent agreement between first 

 readings of whole and sectioned otoliths of canary rockfish 

 caught off Oregon, 1974. 



reader a choice of marginal areas to read, whereas 

 sections do not. 



Additional treatments were applied to otoliths 

 and scales with little success (Table 5). 



TABLE 5. — Treatments applied to otoliths and scales of yellow- 

 tail, canary, and black rockfishes captured off Oregon during 

 1972-75. 



Effect of Deviations of Otolith Readings 

 on Biological Information 



Age Composition 



The frequencies of two independent readings of 

 yellowtail rockfish otoliths made by different 

 readers generally correspond for ages 9-15 (Figure 

 1). Correspondence is lower for younger and older 

 age-groups. The two distributions are approxi- 

 mately normal with means of 12.2 and 12.8 yr, 

 respectively. Figure 5 graphically demonstrates 

 that the means are not significantly different 

 because the 95% confidence intervals for the 

 means overlap. For the two distributions, the 

 standard deviations are similar and the ranges 

 are equal, but the minimum and maximum values 

 disagree by 1 yr (Figure 3). 



Frequencies of age readings for canary rockfish 

 derived from two independent readings by the 

 same person correspond over most of the ranges of 

 ages (Figure 2). Greatest discrepancies occurred 

 at ages 11, 14, and 20. Again the distributions are 

 approximately normal with means of 13.6 and 

 14.2 yr for first and second readings, respectively. 

 The means are not significantly different at the 

 95% level (Figure 5). The standard deviations are 

 similar, while the maximum ages disagree by 2 yr. 



Otolith reading frequencies for two indepen- 

 dent readings by the same person for black 

 rockfish correspond closely for ages 9-12. There is 

 less agreement for other ages (Figure 3). The 



410 



