Stephenson and Hall: Timing of otolith ring formation in marine teleosts from northwestern Australia 



905 



0123456789 1011121314 15161718 0123456789 1011121314 15161718 

 ONDJ FMAMJ J ASONDJFM ONDJ FMAMJ J ASONDJFM 



Month 



Figure 4 



Index of completion at time t, (c,, t), for N. ftircostis for growth-ring categories 2 ,3, 4, 5, 6, and 7-9 sampled 

 between the months 1 October 1993 (^=0) to 31 March 1995 (/=18). The solid circles (•) represent those (c,, 

 t), most likely represented by c"j and lying closest to the function F, and the solid triangles (A) are those 

 most likely represented by c"., and lying closest to the function Fj. The logistic function, P,, indicates the 

 probability that (c,, t) are most likely represented by function Fj. 



that are consistent with those determined in the present 

 study (Table 2). Davis and West ( 1992) found that the time 

 of formation of a new translucent growth-ring on urohyal 

 bones for L. vitta was October, later than the timing found 

 in our study. Sainsbury and Whitelaw (1984) found the 

 marginal increment values on whole otoHths from N. fur- 

 casus had low values in July 1979 and in May 1980 (earlier 

 than observed in the present study for sectioned otoliths) 

 but the sampling reported by Sainsbury and Whitelaw 

 (1984) was very sparse: four sampling times in 1979 and 

 two in 1980. McPherson and Squire (1992) reported that 

 the mean monthly marginal increment of the first two age 

 classes for L. sebae appeared to have a minimum between 

 July and September that is consistent with the present 

 study. 



In our study, the growth zones on L. vitta were generally 

 clearly defined; the opaque zone was easily distinguished 

 from the translucent zone, and there were few discontinui- 

 ties (areas of dissimilar structure or optical density within 

 the growth zone). This clear definition was especially no- 

 ticeable for the outer growth zones of the otolith for older 

 fish which often had very clear dark zones. This finding is 



consistent with the small confidence intervals for this spe- 

 cies, especially in the fish with a greater number of growth 

 rings in their otoliths. The growth rings on L. sp. 3 and N. 

 furcosus had poor contrast and had many discontinuities 

 which made the analysis difficult. For young L. sebae, there 

 were many discontinuities within the growth zones which 

 made locating the translucent zone difficult. For fish with 

 two to four growth rings, low values of the index of comple- 

 tion occurred when t = 7 and also when t = 12 (Fig. 5). For 

 the older L. sebae (where the number of rings is greater 

 than or equal to 12) the wide zone was very dark and by in- 

 creasing the magnification, the marginal increment could 

 be measured relatively easily. The narrow confidence limits 

 for the timing of growth-ring formation for L. sebae are 

 consistent with this explanation but the small number of 

 data points results in less reliable measures in the timing 

 of new ring formation. 



The time of formation of the new growth ring was 

 slightly earlier for L. vitta than for L. sp.3, N. furcosus, or 

 L. sebae. The calculation of an earlier growth-ring forma- 

 tion may be attributed to the more clearly defined trans- 

 lucent zone in L. vitta which may be detectable earlier in 



