serial number by which the release data could be 

 established with certainty. Thus the new returns 

 strongly support the opinion of Mather et al. ( 1972) 

 that the white marlin is much longer lived that was 

 supposed before the work of de Sylva and Davis 

 (1963) and our tagging results were available. It is 

 still impossible, however, to estimate the total life 

 span of the species. 



Mortality 



Since numerous new tag returns have been ob- 

 tained, the indicated mortality rate and the coeffi- 



cient of instantaneous mortality for recaptured 

 white marlin which had been calculated by Mather 

 et al. (1972) from recaptures in groups A-C from 

 releases in 1961-1965 have been calculated from re- 

 captures in the same groups from releases in 

 1961-1967. The same procedures were followed 

 (Table 3, Fig. 5). The new indicated mortality rate 

 is 30% per year, an increase of 3% over the earlier 

 result, with 95% confidence limits of 23% and 36%. 

 The new coefficient of instantaneous total mortal- 

 ity, Z, is 0.35±0. 10, as against the earlier figures of 

 0.32 ±0.17. The addition of the new data has not 

 changed the indicated mortality rate significantly, 

 but has narrowed the confidence limits (14% and 



Table 3. — Summary of recaptures of tagged white mariin, to 15 July 1972, by years of release 

 and months at liberty. Numbers of returns outside of parentheses are for Groups A-C; 

 numbers in parentheses are of Group D. Dashed lines enclose data used for mortality 

 estimates. 



218 



