FISHERY BULLETIN: VOL. 81, NO. 4 



Commerce 1980a-c; Christensen 5 ), and the tilefish 

 fishery has been the most valuable finfish fishery in 

 New Jersey during 1978-80 and New York in 1979- 

 80 (U.S. Department of Commerce 1980a-c; Grimes 

 et al. 1981; Christensen footnote 5). Small fisheries 

 which exploit another stock of tilefish (Katz et aL 

 1983) exist off South Carolina (Creel 1981; Christen- 

 sen footnote 5), southeastern Florida, and in the Gulf 

 of Mexico (Grimes et al. 1980) 



Despite the economic and ecological importance of 

 tilefish, little is known about its life history or the 

 impact of fishing on the Middle Atlantic- Southern 

 New England population (Freeman and Turner foot- 

 note 4; Grimes et aL 1980). The purpose of this paper 

 is to present age, growth, and mortality information 

 on tilefish from that region, to compare size and age 

 structure of the catch of some of the recently active 

 tilefish fisheries, and to examine changes in the size 

 structure of longline catches during 1974-80. 



MATERIALS AND METHODS 



Samples of tilefish from the Middle Atlantic- 

 Southern New England region were obtained from 

 domestic longline and recreational fisheries, Na- 

 tional Marine Fisheries Service (NMFS) trawl 

 surveys, and the foreign trawl fishery. Sample infor- 

 mation included the nature of the sample (random or 

 nonrandom), capture method, location, and date. 

 Length (FL in cm) was recorded for individual fish 

 and often sex and weight (whole and/or eviscerated) 

 were noted as well. It was often impossible 

 to determine the sex of some small tilefish by macro- 

 scopic examination of the gonads so that three sex 

 classifications — male, female, and unknown — were 

 used. No males were identified < 50 cm FL, and most 

 of the fish of unknown sex were <55 cm FL, though 

 some were as large as 71 cm FL. 



After preliminary examination of scales and sec- 

 tions of sagittal otoliths and third dorsal rays, otoliths 

 were selected as primary aging structure. Up to five 

 sections (0.15-0.35 mm thick) were taken from the 

 center of each otolith in the dorsoventral plane using 

 a diamond blade saw. Sections were examined with a 

 dissecting microscope at 1 OX with reflected light and 

 a dark background. The transition from the translu- 

 cent (hyaline) to opaque tissues was most pro- 

 nounced, and was defined as the edge of an annulus. 

 It was usually impossible to follow annuli around an 

 entire section of a tilefish otolith, so that, when the 



5 D. J. Christensen, Northeast Fisheries Center Sandy Hook 

 Laboratory, National Marine Fisheries Service, NOAA, Highlands, 

 NJ 07732, pers. commun., 1982. 



number of rings was in doubt, rings were counted on 

 each side of the sulcus acusticus (Fig. 1). We made 

 measurements to each annulus and to the otolith 

 edge in the medioventral region of the section which 

 passed through or closest to the center of the otolith 

 We used a filar micrometer and recorded distances in 

 ocular micrometer units (one unit = 0.082 mm). 

 Because tilefish otoliths grow allometrically in the 

 medioventral region, measurements were made from 

 the core of the otolith to the furthest point from the 

 core on each annulus and on the edge of the section 

 (Fig. IB). Hayashi (1976a) made similar measure- 

 ments, though in a different plane, on the whole 

 otoliths of the red tilefish, Branchiostegus 

 japonicus japonicus. 



All otoliths of Lopholatilus chamaeleonticeps were 

 read once, and one- third were reexamined. All of the 

 first 120 otoliths were read twice. Close agreement 

 between first and second readings occurred in a sub- 

 sample of 50 from the next 150 otoliths; therefore, 

 routine second readings were discontinued for fish 

 with <10 annuli. Otoliths with 10 or more annuli 

 were assigned an age only after agreement was 

 reached between several counts; when agreement 

 was not achieved, the median number of annuli from 

 at least five counts was used. 



Empirical lengths at age were used in constructing 

 an age- length key; however, a few fish did not form 

 annuli by the end of the usual period of annulus for- 

 mation. To reduce bias which would result from 

 assigning such fish to a younger age, we adopted the 

 following rule: Any fish captured in the 3 mo after the 

 end of the usual annulus formation period with 

 hyaline tissue at the edge of the otolith and a marginal 

 increment at least half as large as the increase in size 

 of the otolith in the previous full year was assigned an 

 age corresponding to its number of rings plus one. 



Least- squares linear regression was used to de- 

 scribe the otolith size (OS): fork length (FL) 

 relationship and the length: weight relationships. The 

 final regression lines were converted to functional 

 regression equations (Ricker 1973). We added a fac- 

 tor to the OS:FL equation which compensated for 

 variation in otolith size at a given fork length. The dis- 

 tance to each annulus was adjusted by the ratio of the 

 average otolith size for fish of the fork length in ques- 

 tion to the observed otolith size for that fish (Bagenal 

 and Tesch 1978). The resulting equation was used to 

 compute back-calculated lengths. 



Analysis of covariance (ANCOVA) was used to 

 compare slopes of regression lines between sexes, 

 and analysis of variance (ANOVA) was used for com- 

 parison of mean marginal increments, mean growth 

 increments, and mean lengths at age. The SAS 



752 



