90 



Fishery Bulletin 94(1). 1996 



Sebastes jordani is an abundant but lightly uti- 

 lized rockfish that occurs along the central Califor- 

 nia coast (Lenarz, 1980). Importantly, it can be 

 readily identified at all early life history stages 

 (Moser et al., 1977). In addition, a detailed growth 

 model for the first six months of life has been previ- 

 ously described (Laidig et al., 1991). Those authors 

 validated a 1:1 correspondence between changes in 

 counts of daily growth increments and age for a de- 

 veloping cohort of pelagic juvenile S. jordani ( age 3- 

 5 mo). 



In this study we examine more closely the accu- 

 racy of S. jordani age estimates derived from the first 

 month of life. In particular, we address the following 

 questions: 1) Are daily increments large enough to 

 be resolved with an optical microscope (Neilson, 

 1992)? 2) Does the Sebastes "extrusion check" 

 (Penney and Evans, 1985; Laidig et al., 1991) actu- 

 ally form at parturition? 3) How variable are age es- 

 timates obtained from different readers? 4 ) How vari- 

 able are individual larvae in their expression of 

 otolith microstructure? and 5) Does larval age influ- 

 ence the reliability of age estimates? 



38 - 



37 



124 



123 

 Longitude (°W) 



12: 



Figure 1 



Map of the study region showing the locations of bongo- 

 net stations. 



Materials and methods 



The primary set of ichthyoplankton samples used in 

 this study was obtained during a cruise aboard the 

 NOAA RV David Starr Jordan conducted from 8-15 

 February 1991. During that six-and-a-half-day pe- 

 riod, 150 bongo-net stations were occupied in the 

 region bounded by lat. 36°30'N and 38"00N, and off- 

 shore by a maximum distance of 130 km ( Fig. 1 1. This 

 region is known to harbor large spawning aggrega- 

 tions of S. jordani during the months of January and 

 February, especially in the vicinities of Pioneer and 

 Ascension Canyons (Lenarz, 1980; MacGregor, 1986; 

 Chess etal., 1988). 



Field and laboratory processing of the samples 

 closely followed prescribed CalCOFI guidelines (i.e. 

 Kramer et al., 1972; Smith and Richardson, 1977). 

 However, because Sebastes larvae generally occur in 

 the upper mixed layer ( Ahlstrom, 1959; Kenchington, 

 1991; senior author's unpubl. data), the maximum 

 amount of wire deployed was 200 m, resulting in a 

 maximum depth fished of 140 m. After retrieving and 

 washing down the 505-/jm mesh nets, the samples 

 were preserved in 80^ ethanol (EtOH), which was 

 changed 24 hours later. In the laboratory the alcohol 

 was changed again, the samples were sorted, and S. 

 jordani larvae were identified and enumerated. 

 Sorted larvae from each of the 150 ethanol-preserved 

 bongo nets were randomly subsampled and their 

 otolith microstructure was examined. Notochord 



length (NL) was measured from each larva in the 

 subsample and the otoliths were extracted and af- 

 fixed to glass slides with clear fingernail polish. 



Individual ages were determined by the methods 

 outlined in Laidig et al. ( 1991 ). In brief, otoliths were 

 viewed whole under oil emersion with a compound 

 microscope that was equipped with a video camera 

 and monitor (e.g. McGowan et al., 1987), producing 

 a working magnification of l,250x. Counts of daily 

 increments were initiated at a distinct check mark 

 that occurred consistently at a radius of -15-19 //m. 

 The mark clearly encircled the otolith core and also 

 a more recent zone of faint incremental growth. 

 Penney and Evans (1985) and Laidig et al. (1991) 

 both observed this feature and inferred that the check 

 was formed at the time of larval extrusion (i.e. par- 

 turition). Otoliths with this mark and no additional 

 increments were therefore given a nominal age of 

 zero. Increments were counted from the mark to the 

 most distal point along the longest growth axis. A 

 digitizer was used to measure the radius (//m) of the 

 otolith at the extrusion check (R () ) and the inner edge 

 of the dark, protein rich D zone of each subsequent 

 daily increment (R p R 2 , R 3 , ..., etc.). During postpro- 

 cessing the width of each increment (W t ) was then 

 calculated by subtraction (i.e. W = R t - R,^)- A se- 

 ries of computer programs automated this procedure 

 (Laidig and Pearson, 1992). 



A confidence code ranging from 1-5 was assigned 

 by the reader at the time each larva was aged. The 



