NOTES Uchiyama and Sampaga: Age estimation and composition of Pseudopentsceros wheelen 



219 



first dorsal spine, midway along the lateral line, and 

 below the posterior end of the dorsal fin. Three sites 

 were midway between the lateral line sites and the dor- 

 sal fin, and three were midway between the lateral line 

 sites and the ventral border. Scales were taken from 

 20 fish of three different body types and from 13 fish, 

 caught in February 1985, with three check marks 

 on their sagittae. After being cleaned, scales were 

 mounted in Euparal under a cover glass on a glass slide 

 or flattened between two glass slides after being air 

 dried. 



Selection of the hard part for ageing was based on 

 the consistency of perceived annual check mark counts 

 and ease of preparation (Table 1). Only the sagittae, 

 vertebral centrum, hypural plate, and dorsal and anal 

 fin spines showed check marks. Counts of check marks 

 from each part were identical (Table 1), indicating these 

 structures could be used for estimating age if their 

 check marks could be verified as annuli. The number 

 of check marks on vertebral centrum, hypural plate, 

 and ttiaxillary was reduced by half upon drying. Soft 

 rays of dorsal, pelvic, pectoral, and anal fins showed 

 no marks that could be interpreted as check marks. 

 Scales were without check marks, and their circuli ap- 

 peared evenly spaced. Of all the structures compared, 

 the sagitta was considered the best structure for 

 estimating the age of pelagic armorhead. 



Counts of daily growth increments 



Sagittae removed at sea were stored in fresh water in 

 vials and refrigerated; those not extracted at sea were 

 frozen within the head and later removed in the labor- 

 atory. Sagittae were sanded lightly on both sides with 

 no. 400 carborundum sandpaper until translucent and 

 etched in a very dilute solution of HCl for 3-5 minutes. 

 The acid was diluted until only 1-2 bubbles per second 

 were formed in reaction to the calcium carbonate of 

 the sagitta. This process was repeated until the incre- 

 ments became visible from the core to the tip of the 

 postrostrum. The sagittae of postlarval specimens were 

 teased from the skull, cleaned, and mounted in Euparal 

 without further processing. To support the coverslip, 

 each sagitta was mounted on a microscope slide in 

 Euparal by using short segments of monofilament line 

 slightly greater in diameter than the thickness of the 

 otolith. Sagittae were examined under a compound 

 microscope at 300 x magnification. 



Increment counts often took a circuitous route be- 

 tween the core and the tip of the postrostrum; the mean 

 of 10 counts was used as the age estimate. Increment 

 counts were routine for sagittae with two check marks 



Reference to trade names does not imply endorsement by the 

 National Marine Fisheries Service, NOAA. 



but were more difficult for those with three check 

 marks. Increment width in translucent zones decreased 

 with increasing number of check marks (8.3 ytva dur- 

 ing year 1, 1.7 yim during year 2, and 1.5 \xm during 

 year 3), and incremental pattern was more variable 

 after year 2. 



Age estimation 



Check marks couhl ndt be validated as annuli by con- 

 ventional means (Beamish and McFarlane 1983). Mean 

 increment counts were used to estimate the hatching 

 time, assuming the first increment formed at or soon 

 after hatching. This technique, though low in accuracy, 

 may suffice when direct validation is unattainable 

 (Gjt^saeter et al. 1984). To verify check marks as an- 

 nuli, mean counts of growth increments for fish caught 

 in February were then compared with counts of check 

 marks, which were multiplied by 365. 



Age composition of catches of pelagic armorhead at 

 Southeast Hancock Seamount was based on check- 

 mark counts (i.e., perceived annuli). Mean FLs between 

 age-sex groups were compared using the Mann- 

 Wliitney paired test, since unequal variances occurred 

 in one sample. To increase sample size of age-sex 

 groups from the Kitakami Maru catch, randomly 

 sampled specimens from Northwest and Southeast 

 Hancock Seamounts were combined. 



Results and discussion 



Increment counts from demersal and oceanic adults 

 and juvenile pelagic armorhead suggested that incre- 

 ments were deposited daily (Tables 2 and 3). A total 

 of 88% of the estimated hatching dates of adults oc- 

 curred within the spawning period, and the other three 

 hatching dates were reasonably close to the spawning 

 period. 



Check marks appeared to be valid annuli based on 

 number of daily growth increments for fish caught in 

 February (Table 2). Sagittae with two check marks had 

 approximately 730 increments (range 728-773, A^ = 5), 

 and those with three check marks had about 1095 in- 

 crements (range 1037-1128, A/^ = 4). Check marks were 

 also verified as annuli by a second but less precise 

 method (Pannella 1971). Partial counts of 365 apparent 

 daily growth increments appeared to coincide closely 

 with the first pair of opaque and translucent zones of 

 sagittae of age-1 and -2 fish. Partial counts of about 

 730 apparent daily growth increments corresponded 

 with the outer margin of the second translucent zone 

 on sagittae of age-2 fish. 



Fish caught in summer appeared to have completed 

 check marks on their sagittae. Sagittae from fish 

 caught in July with two check marks had increment 



