400 



Fishery Bulletin 94(3), 1996 



abandoned because of inclement weather north of the 

 ridge, restricting most sampling to nearshore waters. 

 Stations were located 13, 14, 15, 16, 25, 28, 28, 207, 

 228, and 232 kilometers offshore. Forty-four tows 

 were conducted in a stepped oblique fashion (mean 

 depth=93 m; SD=12 m). Sampling consisted of be- 

 tween one and seven back-to-back replicate tows at 

 each station with a 4-m 2 ring net equipped with 

 0.505-mm mesh. 



During the Hokusei Maru February 1992 (HM-92) 

 survey, paralarvae were collected along the windward 

 Nihoa and Kauai transects, including three stations 

 southwest of the archipelago along the leeward Nihoa 

 transect and a lone station in the lee of Oahu Island. 

 The Kauai transect included stations sampled dur- 

 ing the HM-91 survey. One hundred thirty-one tows 

 were collected at sixteen sampling stations (mean 

 depth=70 m; SD=14 m). Owing to high variability in 

 the catch rates from the HM-91 survey, the number 

 of tows taken at each sampling station was increased 

 from five to between seven and ten back-to-back rep- 

 licate tows. Stations along the Kauai transect were 

 located 11, 41, 142, 235, 337, 471, 585, and 713 kilo- 

 meters offshore. Nihoa transect stations were located 

 26, 181, and 269 kilometers windward, and 25, 126, 

 and 256 kilometers leeward of the archipelago. Four 

 scheduled stations along the windward Nihoa 

 transect were canceled because of inclement weather. 



During the RV Townsend Cromwell March-April 

 1992 ( TC-92 ) survey, four to eight replicate tows were 

 conducted at fourteen stations along the windward 

 Nihoa and Oahu transects. One hundred tows (96 

 with a 2-m 2 ring; 4 with a 4-m 2 ring) were conducted 

 in a stepped oblique fashion (mean depth=94 m; 

 SD=13 m). The net carried a Wildlife Computers 

 time-depth recorder. Stations were located between 

 52 and 1,161 kilometers offshore. 



During the Hokusei Maru February 1993 (HM-93) 

 survey, paralarvae were collected along the same two 

 transects, northeast of the Hawaiian Island chain, 

 that had been sampled in February 1992, including 

 three stations southwest of the archipelago (leeward 

 Nihoa transect) and a lone nearshore test station 

 (only two tows conducted) northeast of Oahu Island. 

 Ninety-five tows (mean depth=92 m; SD=23 m) were 

 collected at twelve sampling stations. Seven to ten back- 

 to-back replicate tows were taken at each station. 



Specimens from the Hokusei Maru surveys were 

 fixed in 4% formalin and preserved in 50% isopropyl 

 alcohol within 3-6 hours to prevent possible acidic 

 degradation of the calcareous statoliths. TC-91 sur- 

 vey samples were fixed in 10% formalin and pre- 

 served in 50% isopropyl alcohol. TC-92 survey 

 samples were fixed in 6% formalin and preserved in 

 50% isopropyl alcohol. Statoliths from specimens 



stored in formalin longer than six hours darkened 

 and proved unreadable. 



Paralarvae of the family Ommastrephidae were 

 identified to genus and species level according to fig- 

 ures and descriptions of Hawaiian ommastrephids 

 published by Young and Hirota ( 1990) and Harman 

 and Young ( 1985). The dorsal mantle length of each O. 

 bartramii paralarva was measured to the nearest 0.1 

 mm with an ocular micrometer in a stereomicroscope. 



Statoliths were removed from the paralarvae with 

 fine dissecting forceps under a stereomicroscope. First, 

 the paralarval funnel was removed, revealing the ovate 

 statocyst. Statoliths were clearly visible within the sta- 

 tocyst as paired, white, opaque structures. The excised 

 statocyst was transferred to a microslide, where it was 

 gently pulled apart, permitting one of the statoliths to 

 fall onto the slide. The statocyst with the remaining 

 statolith inside was then transferred to a separate slide 

 where the statolith was similarly removed. Statoliths 

 were rinsed in distilled water and air-dried before 

 analysis. When dried, statoliths were embedded in ther- 

 mal plastic by heating the plastic to 77°C. 



Statoliths from 85 animals (mantle length: 1.0— 

 4.7 mm) were used for growth increment analysis. 

 Paralarvae used for statolith examination were col- 

 lected during the HM-91, HM-92, and TC-92 surveys. 

 Growth increments on O. bartramii statoliths con- 

 sisted of two components, a broad, translucent ring 

 and a narrow, dark ring. Using a Zeiss compound 

 microscope (400x), increments were counted from the 

 nucleus to the outer margin of the dorsal end of the 

 lateral dome. Examinations were made with trans- 

 mitted light. Counts of each statolith were conducted 

 on three separate occasions and averaged for a mean 

 increment value. Increments in the outermost por- 

 tion of the lateral dome and the innermost portion 

 near the nucleus could not be counted in some cases; 

 therefore the number of increments on those portions 

 was estimated by extrapolation. 



Data treatment 



Age for each of the 85 paralarvae used for growth 

 increment analysis was determined from the mean 

 of the three counts of statolith growth increments. 

 The exponential equation y=ae bx was fitted to the 

 data for the mantle-length-increment-count relation- 

 ship. Age estimates of all other paralarvae were based 

 on this relationship curve. 



Estimated ocean current speeds and 

 directions 



Patterns of ocean currents near the Hawaiian Ar- 

 chipelago were defined by the trajectories of near- 



