360 



Fishery Bulletin 90(2). 1992 



The correlation coefficient (r) for this relationship in 

 C. arel is 0.6043 (P< 0.001) and is statistically signif- 

 icant; in C. lida, it is 0.2579 (P>0.05) and is not 

 statistically significant. 



Fecundity against TW showed a linear relationship 

 (Fig. 16), and equations for the transformed data are, 



a arel logF = 3.5907 + 0.4995 logTW 

 C. lida logF = 3.5536 + 0.6108 logTW. 



The correlation coefficient (r) for this relationship in 

 C. arel is 0.6345 and is highly significant (P<0.001); 

 in C. lida, the correlation coefficient of 0.2302 (P>0.05) 

 did not indicate a significant correlation between these 

 two variables. 



Ovary length showed a straight-line relationship with 

 fecundity (Fig. 16). In logarithmic form, the relation- 

 ships between F and OL can be expressed as follows; 



a arel logF = 1.8472 + 1.3693 logOL 

 C. lida logF = 0.3206 + 2.2630 logOL. 



The correlation coefficient in C. arel is 0.6632 (P< 

 0.001) and is statistically significant; in C. lida, it is 

 0.4990 {P< 0.05), showing a high degree of correlation. 

 Fecundity plotted against OW showed a linear rela- 

 tionship (Fig. 16) and equations for these two vari- 

 ables are, 



C. arel logF = 2.1858 + 0.7050 logOW 

 C. lida logF = 1.3909 + 1.0038 logOW. 



The correlation coefficient in C. arel is 0.7729 (P< 



0.001), indicating a high degree of correlation between 

 these two variables. In C. lida, the correlation coeffi- 

 cient of 0.8606 (P< 0.001) is highly significant. 



In the present study, the exponential value (b) for 

 total length-fecundity was higher than for total weight- 

 fecundity. Similarly, the b value for ovary length- 

 fecimdity was higher than for ovary weight-fecundity. 



Sex ratio The sex ratio was about 1:1 for both spe- 

 cies (Table 18, Fig. 14). However, the ratio varied in 

 monthly samples, and chi-square values showed a 

 significant deviation from the expected 1:1 ratio for 

 3 months (February, September, and November) in 

 C. arel, and during February-April, September, and 

 November in C lida. Since data were pooled for one 

 year, the chi-square value conformed to the expected 

 1:1 ratio in C arel; whereas in C. lida, it deviated 

 significantly from the expected 1:1 ratio. The devia- 

 tion may be due to multiple testing. 



Discussion 



Feeding ecology 



Cynoglossus arel and C. lida feed predominantly on 

 polychaetes and crustaceans, followed by other phyla 

 such as molluscs, echinoderms, and coelenterates. 

 These similarities in diets indicate common feeding 

 strategies within the tonguefishes and soles (Seshappa 

 and Bhimachar 1955, Kuthalingam 1957, de Groot 

 1971, Braber and de Groot 1973ab, Stickney 1976, 

 Pearcy and Hancock 1978, Langton and Bowman 1981, 

 Wakabara et al. 1982, Langton 1983, Honda 1984). 



