Table 6. — Results of analysis of covariance of morphometric character as a function of 

 eye-fork length. The statistical test is whether the relation is significantly different among 

 areas, (n.s. = not significant; * = 5% significance level; ** = 1% significance level). 



Character 



Weight 



Ma.ximum body depth 

 Depth at vent 

 Length of pectoral fin 

 Length of pelvic fin 

 Snout to mandible length 

 Snout to eye length 

 Dorsal fin height 

 Length of maxillary 



eluding results for weight-length relations, results of 

 the statistical test of ci =0 indicate that most of the 

 fl's are significantly different from zero. This sug- 

 gests that growth of the body parts is allometric, or 

 the parts do not grow as a constant proportion to 

 body size, which is characteristic for most body 

 parts of fishes (Martin. 1949). 



Analysis of covariance was performed to test 

 whether the regressions differed between sampling 

 locations. No significant differences were found 



(Table 6). Samples from Buena Vista and Mazatlan 

 were therefore pooled and the regressions were re- 

 calculated (Table 7). 



SAILFISH 



A total of 81 1 sailfish was sampled at Buena Vista 

 and Mazatlan. Sampling at Buena Vista was in 

 1967-70 and at Mazatlan. only in 1967-69. More 

 fish, however, were sampled at Mazatlan than at 



Table 7. — Regression of morphometric character on eye-fork length (cm) for pooled (loca- 

 tions and sexes) samples of blue marlin and sailfish from the eastern Pacific. Weight-length 

 relation is based on log transformed data (log Y = a + h]ogX): all other relations are based 

 on untransformed data {Y = a + hX). 



113 



