FISHERY BULLETIN: VOL. 79, NO. 1 



their tetracycline-otolith method. Aikawa and 

 Kato (1938), Nose et al. (1957), Yabuta et al. (1960), 

 Tan et al. (1965), and Shabotiniets (1968) inter- 

 preted marks on scales, dorsal spines, and the cen- 

 trum of vertebrae to estimate age and growth. 

 These studies were performed on commercial- 

 sized fish ( > 2 kg) ; growth during early life (< 2 kg) 

 has yet to be examined. 



Pannella's reports (1971, 1974) provided cir- 

 cumstantial evidence that the smallest discern- 

 ible growth increments in the sagittae (otoliths) of 

 fish are deposited daily. More recent studies pro- 

 vide direct evidence that these growi;h increments 

 are diel phenomena in sagittae of temperate 

 (Brothers et al. 1976; Taubert and Coble 1977; 

 Barkman 1978) and tropical (Struhsaker and 

 Uchiyama 1976; Wild and Foreman 1980) species 

 of teleosts. In the study of the short-lived en- 

 graulid Stolephorus purpureas, the information 

 gained from the reading of sagittae was utilized in 

 the construction of a growth curve for the first 190 

 d after yolk-sac absorption (full life cycle) 

 (Struhsaker and Uchiyama 1976). In the present 

 paper, growth curves are presented for central 

 Pacific skipjack tuna to an age of about 3 yr, based 

 on a sample of 51 fish, and for yellowfin tuna to 

 about 2 yr, based on 14 fish. Counts on sagittae 

 from 20 skipjack tuna from the eastern Pacific and 

 5 skipjack tuna from Papua New Guinea are also 

 given. The results are discussed in relation to ear- 

 lier age and growth studies of these species. 



METHODS 

 Otolith Preparation and Counting 



The central Pacific skipjack and yellowfin tunas 

 were caught in the vicinity of the Line Islands and 

 Hawaii. All specimens >20 cm FL (fork length) 

 are samples taken from commercial fisheries or 

 caught by trolling. Specimens <20 cm FL are from 

 stomach contents of troll-caught skipjack tuna 

 and regurgitations of a seabird, Sula sp., after it 

 landed on the deck of a research vessel. 



Juvenile skipjack tuna from stomach contents 

 were identified by vertebral counts and skeletal 

 characters given by Godsil and Byers (1944) and 

 Gibbs and Collette (1967). In one case, only the 

 anterior portion of a juvenile skipjack tuna was 

 collected; standard length (SL) was estimated 

 from the length of the precaudal vertebrae using 

 the equation given by Yoshida (1971). A small (7.0 

 cm FL) yellowfin tuna specimen was tentatively 



identified on the basis of skeletal characters given 

 by Matsumoto et al. (1972) and descriptions of 

 Thunnus livers by Godsil and Byers (1944) and 

 Gibbs and Collette (1967). 



The caudal rays were missing from most tuna 

 specimens collected from stomachs. Fork lengths 

 were estimated by increasing standard lengths by 

 3.3% (Matsumoto^). 



Heads from which the sagittae were not im- 

 mediately removed after collection were frozen or 

 preserved in 75% isopropanol. 



In tunas <100 cm FL, we obtained the sagittae 

 by splitting or cutting the skull along the sagittal 

 plane and teasing them from the semicircular 

 canals. With experience, the cut could be made 

 without damaging either of the sagittae. We 

 cleaned the sagittae by teasing or brushing off the 

 sacculus and nerve endings. Sagittae that were 

 not mounted immediately were stored in distilled 

 water or 40% isopropanol, but were then mounted 

 within a year. 



After removal and cleaning, sagittae from tunas 

 >45 cm FL were etched in a 1% solution of HCl for 

 3-5 min. The otoliths were then rinsed with sev- 

 eral changes of water, mounted in Euparal,^ and, 

 in most cases, permitted to clear for 4 wk or more. 

 Short lengths of monofilament line prevented con- 

 tact of the otolith with the glass cover slip. 



Increment counts were made from the core,^ the 

 center of the nucleus, to the tips of the rostrum, 

 antirostrum, and postrostrum (terminology of 

 Messieh 1972) for most sagittae. On sagittae from 

 fish >60 cm FL, counts were made only to the 

 rostrum and postrostrum. When the edge of the 

 rostrum and postrostrum was fringed with irregu- 

 lar projections, counts were made on the projec- 

 tions leading to the point. Specimens were 

 examined at magnifications of 200-800 x. A mi- 

 croscope with a zoom feature was found to be use- 

 ful, as the width of daily increments varied. 



Counting increments in whole mounted otoliths 

 becomes progressively more difficult with increas- 

 ing specimen size. Experience is best obtained by 

 beginning with otoliths from young fish and pro- 

 gressing through older fish. The initial counts en- 



^W. M. Matsumoto, Southwest Fisheries Center Honolulu 

 Laboratory, National Marine Fisheries Service, NOAA, Hon- 

 olulu, HI 96812, pers. commun. October 1974. 



''Reference to trade names does not imply endorsement by the 

 National Marine Fisheries Service, NOAA. 



''Terminology agreed upon at the Otolith Workshop, Scripps 

 Institution of Oceanography and Southwest Fisheries Center, 

 National Marine Fisheries Service, NOAA, La Jolla, Calif , 12-16 

 July 1976. 



152 



