KISHKRY Bl'LLETIN: VOL. 81. NO 4 



frequency data) decreased in 1974 to 1980. The 1974 

 data indicated that complete recruitment occurred 

 after 8 1-85 cm. By 1976 the peak frequency occurred 

 at 6 1-65 cm, and this continued to be the case in 1977 

 and nearly so in 1978 in the Hudson Canyon area. In 

 the Southern New England area, the lowest modal 

 size occurred during 1978 at 51-55 cm FL. In suc- 

 ceeding years the modal size was generally 56-60 cm 

 FL for both areas. 



Strong modes at successively larger sizes in con- 

 secutive years were apparent in the longline length- 

 frequency data, suggesting the presence of strong 

 year classes and variable recruitment. The presence 

 of such modes in both fishing areas indicated that 

 they were not simply a result of discovering new con- 

 centrations of tilefish as the fishery expanded. Par- 

 ticularly noticeable was a mode at 41-45 cm FL in 

 1977, 50-55 cmFL in 1978, and 56-60 cm FL in 1979. 

 These sizes were similar to our empirical lengths at 

 age in 1978 of 4-, 5-, and 6-yr-old fish suggesting that 

 these modes represented the 1973 year class. 



Age Structure and Mortality 



Age composition of the longline, recreational, and 

 foreign trawl fisheries was quite different in 1978. 

 Longline landings were dominated by 7 and 8 yr olds 

 which represented 24 and 25% of the catch, respec- 

 tively, and 90% of the catch were 4-9 yr olds (Fig. 9). 

 The recreational catch was comprised mainly of 5 yr 

 oids (32%), and 987 of the catch were 4-9 yr olds. 

 Four-yr-old tilefish accounted for 33% of the foreign 

 trawl catches; 72% of the catches were 3-5 yr olds, 

 and 9 1 7 were 3-7 yr olds. Age at full recruitment was 

 clearly different for each fishery — age 9 for longline, 

 6 for recreational, and 5 for the foreign trawl. The dif- 

 ferences in length frequencies of longline catches in 

 the Hudson Canyon and Southern New England 

 areas resulted in higher percentages of 7 yr olds in the 

 former area and more 4-6 yr olds in the latter. 



15 



10- 



0) 



o 



5- 



• • 



5 10 15 20 25 30 35 



Age (years) 



Ficure 9.— Catch curve for tilefish caught by longline during 1978 

 in the Middle Atlantic-Southern New England region. Regression 

 lines from data for fish age 9-20 yr and 9-35 yr are shown. 



Total mortality rates (Z) computed for the longline 

 and foreign trawl fisheries were found to be similar. 

 ANCOVA revealed no significant differences be- 

 tween sex-specific catch curves in total mortality 

 rates in the 1978 longline fishery (P > 0.05), thus Z 

 was computed for all tilefish combined. The number 

 at age in the longline fishery (Fig. 9) was typical of 

 both fisheries in that there were numerous fish at the 

 younger ages (<10 yr) and few fish between 10 and 

 35 yr. This resulted in a concave catch curve indicat- 

 ing either a lower rate of mortality for the oldest fish, 

 or that those fish had not been subjected to fishing 

 pressure for their entire lives, or both. The total mor- 

 tality estimate from the longline data was 0.17 (r 2 = 

 0.416) for all ages; however, the effect of the cur- 

 vilinearity in the catch curve was considered signifi- 

 cant, and a more representative age range was 

 selected. The instantaneous rate of total mortality 

 was reestimated to be 0.46 for ages 9-20 from the 

 longline fishery (the number at age 1 1 was con- 

 sidered an outlier and not included, r 2 = 0.52). The 

 rate of total mortality for the foreign trawl fishery was 

 estimated to be 0.60 for ages 5-8 (r 2 = 0.84). 



DISCUSSION 

 Growth 



All branchiostegids studied to date exhibit differ- 

 ences in growth rates between sexes with the fe- 

 males growing more slowly than males. Lopholatilus 

 chamaeleonticeps, the largest and longest lived 

 branchiostegid, shows the most pronounced dif- 

 ference in this regard with males reaching 105-115 

 cm FL in 20-26 yr and females ranging from 85 to 95 

 cm FL in29-35 yr. Hayashi (1976b) reported that the 

 average male Bronchiostegus japonicus japonicus 

 from the East China Sea was 32 cm TL (reported as 

 body length) at 8 yr while females averaged only 29 

 cm TL at 8 yr. Ross (1978) demonstrated similar, 

 though proportionally less pronounced, differences 

 for Caulolatilus microps from North Carolina and 

 South Carolina. Differences in mean back-calculated 

 length at age between the sexes during the oldest 

 ages were on the order of 15-207 of the average 

 male' s length forL. chamaeleonticeps (this study) ,10- 

 127c for R. japonicus japonicus (Hayashi 1976b), and 

 from —1 to 57 for C. microps (Ross 1978). 



The divergence of growth rates in branchiostegids 

 is correlated with sexual maturation of females. Pro- 

 nounced differences in growth between the sexes inL. 

 chamaeleonticeps was evident at 5 yr (Fig. 6) when the 

 average female was slightly smaller (51 cm empirical 

 FL) than the observed size of maturation (52-57 cm 



760 



