Radtke and Hourigan Age and growth of Nototheniops nudifrons 



559 



feasible to enumerate all increments. Individual sec- 

 tions were etched for different lengths of time, with 

 15-20 minute etching times showing the inner incre- 

 ments most clearly, and shorter etching periods reveal- 

 ing the outer increments. Landmark scratches were 

 placed on each section with an insect needle to follow 

 increments uncovered by different etching times. The 

 otolith was etched in steps of 1 minute and viewed with 

 the SEM after each etching to follow the progression 

 of the smallest increments found. Examination of the 

 three-dimensional microstructure of the left sagitta 

 demonstrated that the midtransverse plane was the 

 best plane for routine sectioning. 



Validation of otolith ageing technique 



Microincrements are generally assumed to be deposited 

 daily (Pannella 1971, 1974; Radtke and Dean 1982; 

 Radtke and Targett 1984). To test this, individual fish 

 were given a single intramuscular injection of either 

 tetracycline hydrochloride or acetazolamide at 0.025 

 mg/g body weight. Tetracycline is incorporated into the 

 otolith and, viewed under a compound microscope with 

 filtered reflected wavelengths of 700-800 nanometers, 

 appears as a fluorescent band inside the otolith. Tetra- 

 cycline marks are not visible by SEM. Acetazolamide 

 is a carbonic anhydrase inhibitor which temporarily in- 

 hibits the calcification processes in the otolith resulting 

 in a disruption band which can serve as a reference 

 location for SEM study (Mugiya and Muramatsu 1982, 

 Radtke unpubl. data). 



Left sagittae from the fish injected with tetracycline 

 were embedded in epoxy-casing resin and serially sec- 

 tioned using a low-speed saw. The sections were 

 polished as above, placed in nonfluorescent immersion 

 oil, and examined under reflected ultraviolet light 

 through a compound microscope. Otoliths with aceta- 

 zolamide marks were prepared for SEM analysis as 

 described above. The number of microincrements from 

 the reference mark to the margin of the otolith was 

 compared with the number of days between injection 

 and the day that the fish was sacrificed. 



Relationship of otolith dimensions and fish age 



In an effort to determine other predictors of age in A^. 

 nudifrons, the data generated from otolith and fish 

 measurements were applied to the multiple regression 

 model: 



Age = a + bii'i + 60X2 + ^: 



'3 -'3 



+ h,x., 



In this model, age in years was determined by count- 

 ing microincrements, a = intercept, h = regression 

 coefficients, x = variables. The data were checked for 



normality, and the multiple regressions were deter- 

 mined in a stepwise fashion with the inclusion level for 

 variables set at p = 0.05. As a measure of the variance 

 associated with the otolith morphometric procedure, 

 the percent of the total variance component due to 

 variance lietween age estimates from the left and right 

 sagittae of the same fish was calculated. Less than 5% 

 of the total variance was due to variance between sagit- 

 tae of the same fish (Table 1). 



Results 



A total of 216 Nototheniops nudifrons were collected 

 in trawls off Low Island. Five were caught during six 

 deep trawls in 100-110 m depth, with the remaining 

 fish caught in four shallow trawls at depths of 54-80 

 m. Fish captured in trawls ranged in size from 70 to 

 149 mm SL. More females (A^ = 127) were collected 

 than males (N = 74), and the modal size of females (100 

 mm SL) was slightly larger than that of males (90 mm 

 SL; Fig. 1). Three smaller juveniles (each 45 mm SL) 

 were collected by hand in 10 m depth at Arthur Harbor. 



Length-weight relationships 



Length-weight relationships of males and females are 

 shown in Figiu'e 2. Separate regression equations were 

 calculated for males and females using a gonad-free 

 body weight. Weight was approximately a cubic func- 

 tion of length, indicating nearly isometric growth. The 

 length-weight relationship of the fish conformed to the 

 power curve model Weight = o(SL)*. An analysis of 

 covariance indicated that the slopes of the length- 

 weight relationships for the sexes were not significant- 

 ly different (df = 1, 203; F= 2.30, 0.05<p<0.25). The 

 conversion of standard length (SL in mm) to total 

 length (TL in mm) was: 



TL = 1.136(SL) -I- 1.529; (Af= 216; /?- = 0.99). 



