Connaughton and Taylor: Seasonal cycles in the sonic muscles of Cynoscion regalis 



699 



from C. V. Sullivan (Dept. Zoology, North Carolina 

 State University, Raleigh, North Carolina). This an- 

 tiserum cross-reacted less than 2% with testoster- 

 one (Hourigan et al., 1991). 



Extraction efficiencies for the 11-ketotestosterone 

 assay, determined by extraction of samples spiked 

 with a known amount of radiolabelled 11-ketotes- 

 tosterone, were always greater than 85%. RIA par- 

 allelism was determined by measurement of various 

 equivalents of plasma from a single plasma pool. The 

 results were parallel to the standard curve over a 

 range of 1-20 pL plasma for the testosterone assay 

 and 25-250 pL plasma for the 11-ketotestosterone 

 assay (sample sizes for the assays were 5 and 50 pL, 

 respectively). When a range of steroid concentrations 

 was added to plasma pool samples of known hormone 

 concentration, the quantities of spiked steroid recov- 

 ered were not significantly different from the quanti- 

 ties added for either assay. The intra- and inter-assay 

 coefficients of variation were, respectively, 10.3% and 

 18.5% for the testosterone assay and 7.4% and 17.1% 

 for the 11-ketotestosterone assay. 



Statistical analyses 



Specimens with a TL outside a 15-cm range around 

 the sample mean for each year were not included in 

 this analysis. This limited size range was used for 

 two reasons: first, to alleviate the possible effects of 

 allometric growth on the indices used to present the 

 data; and second, to remain within the size range of 

 two- and three-year-old weakfish used by Villoso 

 ( 1989). The calculated indices, such as GSI, were used 

 only to display the data. Statistical analyses on tes- 

 tis weight and sonic muscle morphometric measure- 

 ments were conducted by using analyses of covari- 

 ance with TW or TL of the specimen considered as a 

 covariate. Bilateral comparisons of sonic muscle 

 morphometries were made by using a paired r-test. 

 Statistical analyses of plasma androgen levels were 

 accomplished by using a one-way analysis of vari- 

 ance. The a level for these analyses was 0.05. 



Results 



A difference was noted between 1990 and 1991 sur- 

 face water temperatures, which rose more rapidly in 

 1991 (Fig. 1). Surface temperatures reached 26°C by 

 late May in 1991 but not until late June in 1990. 

 The apparent result of these temperature differences 

 was a two-week difference in the course of events 

 across the collecting period so that 1991 trends be- 

 gan earlier than those in 1990. 



g> a 

 <D _ 



S <2 



55 ?l 



Apr 26 



Sep 23 



Figure 2 



Sonic muscle thickness index (SMTI) and sonic 

 muscle-somatic index (SMSI) of the weakfish, Cyno- 

 scion regalis, plotted across sampling date for the 

 1990 and 1991 collecting seasons. Points are means 

 (rc=3-10 fish) ± one standard error of the mean. 



Sonic muscles were bilaterally symmetrical; we 

 were unable to detect significant differences between 

 the two sides in length, thickness, weight, or color in 

 1990 or 1991. Sonic muscle width on the left side 

 was significantly longer than that on the right side 

 in 1990, but this trend was not repeated in 1991. 



Mean sonic muscle thickness changed significantly 

 across the collecting season in both years. During 

 both collecting periods, SMTI climbed to a peak of 

 approximately 1.5% of TL by late June before steadily 

 decreasing to post-spawning values of approximately 

 0.6%' of TL (Fig. 2). Concurrently, there was a sig- 

 nificant seasonal change in total sonic muscle weight 

 in both 1990 and 1991. SMSI values rose to between 



