BOEHLERT and SASAKI: PELAGIC BIOGEOGRAPHY OF THE ARMORHEAD 



cess of 1.5 years in early summer (Uchiyama and 

 Sampaga fn. 3) and 2 years in winter, suggesting 

 that some recruitment occurs in spring. 



In most fisheries, temporal recruitment patterns 

 can be discerned from length-frequency analyses, 

 because the smaller recruits represent a larger pro- 

 portion of the population during the season of 

 recruitment. For the armorhead, no increase in 

 smaller size classes is apparent in monthly length- 

 frequency samples. This may be due to cessation of 

 somatic growth after recruitment to the seamount 

 (Humphreys and Tagami 1986; Uchiyama and Sam- 

 paga fn. 3). Early recruits, with their deeper bodies 

 and greater fat content, differ morphologically from 

 longer seamount residents. The transition from 

 so-called "fat" to "lean" morphotypes occurs in 

 association with development of gonads (Humphreys 

 et al. in press). Thus, an index of morphological 

 change can be used to detect recruitment patterns, 

 much the same as length frequencies can be used 

 for other species. Condition factor is frequently used 

 to assess "fatness" of fish; it is normally expressed 

 as weight divided by length to a power (typically 3) 

 multiplied by some scaling factor (Ricker 1975). For 

 armorhead, there is a marked change in condition 

 factor with morphological change; newly recruiting 

 fish have high condition factors relative to inter- 



mediate or lean fish. By considering the monthly 

 proportion of fish above an arbitrary value of con- 

 dition factor, we can estimate the influx of new 

 recruits to the seamounts. During May 1972- 

 December 1973, the major influx of new recruits 

 with a high condition factor occurred in April-May 

 1973 (Fig. 5). The data for 1972 are incomplete, but 

 some recruitment apparently occurred in August 

 and September (Fig. 5). Differences in seasonality 

 of recruitment between 1972 and 1973 may be 

 indicative of interannual variability in temporal 

 recruitment patterns. 



An influx of fish to the seamounts may also be 

 reflected in the catch per unit effort (CPUE), an 

 index of stock abundance. Monthly averages of 

 CPUE (in metric tons per hour) by Japanese 

 trawlers for armorhead at the SE-NHR seamounts 

 were highest in March and April (Sasaki 1986). 

 Overall armorhead CPUE decreased continuously 

 from 54.1 t/h in 1972 to only 0.3 t/h in 1982 (Sasaki 

 1986) and further decreased to 0.06-0.07 t/h in 

 1983-85 (T. Sasaki unpubl. data). The estimated 

 CPUE in 1986, however, was 0.31 t/h, an increase 

 of four or five times that in the previous year. The 

 CPUE of trawlers operating in May-Septem.ber 

 1986 was highest in May, decreased in June and 

 July, but increased slightly in August and Septem- 



40 



35- 



30- 



H 25 



z 



HI 



O 20- 



oc 



UJ 



°- 15H 



10 



5 6 7 8 9 10 11 12 1 2 3 4 5 6 7 8 9 10 11 12 



1972 



MONTH 



1973 



Figure 5.— Seasonality of the proportion of armorhead, Pseudopentaceros wheeleri, with 

 condition factors >2.0 from four seamounts of the northern Hawaiian Ridge (Southeast 

 Hancock, Northwest Hancock, C-H, and Colahan) for 1972-73. Squares, males; triangles, 

 females. Data are based upon 2,104 males and 1,704 females. 



461 



