GRIMES ET AL.: REPRODUCTIVE BIOLOGY OF TILEFISH 



uniform ova size (developmental stage) distribution 

 among anterior to posterior ovarian lobe locations, 

 as has been demonstrated for Lopholatilus chamae- 

 leonticeps from the South Atlantic Bight (Erickson 

 et al. 1985). 



To determine if the liver was being utilized to store 

 energy in the form of fat reserves to be used in 

 gonad maturation, we calculated a hepatosomatic 

 index (HSI) = liver weight (g)/gutted body weight 

 (g) X 100 (Htun-Han 1978). 



Ovaries used to estimate fecundity ( = ovarian egg 

 count of Gale and Deutsch (1985)) were preserved 

 in modified Gilson's fixative (Bagenal and Braum 

 1978). The ovarian tunic was removed and washed 

 free of adhering ova. Developing ova were separated 

 from folicular material and most oogonia by wash- 

 ings under a stream of water. Based upon initiation 

 of yolk accumulation, all oocytes with diameters 

 >0.15 mm were included in ovarian egg counts. 

 Each sample of developing ova was diluted in water 

 and stirred, then at least two subsamples pipetted; 

 each subsample was placed in a 6 x 6 cm gridded 

 Petri dish for counting. Ova were counted in six ran- 

 domly selected grid squares, and the average num- 

 ber of ova in the six squares was then adjusted to 

 the total subsample count by multiplying the aver- 

 age by the total number of grid squares in the dish. 

 The sample and subsamples were oven dried at 40°C 

 for at least 24 hours and weighed to the nearest 

 0.001 g on a Mettler^ balance. Fecundity (total 

 ovarian egg count) was estimated as the number in 

 the subsample multiplied by sample weight divided 

 by subsample weight. 



Predictive equations of fecundity from length and 

 weight were fit using least squares regression and 

 converted to functional regressions (Ricker 1973). 

 Fecundity was separately regressed on FL and 

 gutted weight using all possible combinations of un- 

 transformed log and semi-log models. We inspected 

 residuals, plots and coefficients of determination to 

 evaluate fits. 



RESULTS 



Gonad Structure and Sex Determination 



Males smaller than 600 mm FL were difficult to 

 sex by gross gonadal structure because testes were 

 small and undeveloped. However, in females larger 

 than about 400 mm FL ovaries were sufficiently 

 developed to visually determine sex easily. There- 



^Reference to trade names does not imply endorsement by the 

 National Marine Fisheries Service, NOAA. 



fore, we histologically examined gonadal tissues of 

 155 fish (50 males, 545-814 mm FL; 52 females, 

 241-678 mm FL; and 53 juveniles, 146-400 mm FL). 

 Based upon gonad microstructure, seven fish be- 

 tween 467 and 592 mm FL that were macroscop- 

 ically unsexable were males. All juveniles and un- 

 sexed fish <400 mm FL (79 fish 121-400 mm FL) 

 appeared to be females, i.e., had cell structure con- 

 sisting of only oogonia. These cells had slightly aci- 

 dophilic cytoplasm and distinct, but irregular, cell 

 membranes. The nuclear membrane was less ob- 

 vious because it was masked by basophylic nuclear 

 material located peripherally. 



The paired ovaries of adult L. chamaeleonticeps 

 are suspended below the swimbladder from the dor- 

 sal body wall in the extreme posterior of the body 

 cavity. Ovarian microstructure and development are 

 identical to Caulolatilus microps (Ross and Merriner 

 1983) and L. chamaeleonticeps (Erickson et al. 1985) 

 from the South Atlantic Bight and are described in 

 detail by Idelberger (1985). 



The testes of tilefish are solid, smooth textured, 

 and more elongate than ovaries. In males smaller 

 than about 65 cm FL, testes were not obviously 

 lobed and were pinkish in color. Only in large adult 

 males were testes creamy white and heavily lobed. 

 Microstructure of the testes was typical of teleosts 

 (Hoffman 1963; Smith 1965; Hyder 1969) and was 

 identical to C. microps (Ross and Merriner 1983), 

 and, like the ovary structure, was described in detail 

 by Idelberger (1985). Two males (604 and 609 mm 

 FL) had slight amounts of ovarian tissue within the 

 testicular mass (Fig. 1). That ovarian tissue con- 

 sisted of several resting or previtellogenic (peri- 

 nucleolar) oocytes located in testicular mass with 

 definite spermatogenic tubules and crypt structure. 



Sexual Dimorphism 



The predorsal adipose flap is sexually dimorphic 

 in adult tilefish and can easily be used to determine 

 sex in animals larger than about 70 cm FL. The size 

 of the predorsal flap was significantly larger in 

 males than in females older than age 5 years and 

 larger than 65 cm FL (Table 1). It was not possible 

 to compare predorsal flap sizes of the largest males 

 to those of females, because females do not grow 

 as large as males (Turner et al. 1983; Harris and 

 Grossman 1985). 



Sexual Maturity 



We determined the general pattern of age and size 

 of sexual maturity using both visual staging and 



747 



