Le GUEN and SAKAGAWA : APPARENT GROWTH OF YELLOWFIN TUNA 



POINTE-NOIRE 



Figure 1. 



-Area off Africa where the surface fishery 

 for yellowfin tuna operates. 



ployed the relation, log Lf = 0.273 + 1.175 logLd 

 to convert samples with predorsal length in centi- 

 meters (Ld) to fork length in centimeters (Lf) . 

 This relation is based on 508 observations and 

 differs from Lf = (3.624 + 0.212 La)-, which 

 was employed by Poinsard (1969). It has a 

 slightly better correlation coefficient (r — 

 0.9943) than Poinsard's equation (r = 0.9940) 

 (Lenarz, 1971).' Calculated fork lengths based 

 on either equation are accurate only to 1-4 cm. 

 Monthly length-frequency distributions were 

 tabulated by 4-cm-fork-length groups for sam- 

 ples from each region. Modes were identified 

 and assumed to represent age classes within 

 which lengths were normally distributed. Nor- 

 mal distributions were then fitted to the length 

 frequencies of samples in which two or more 

 modes were present, and the mean length of each 

 age class was estimated (Table 2). A computer 

 program for separating size classes in a mix- 

 ture that was written and described by Hassel- 



* Lenarz, W. H. 1971. Length-weight relations for 

 five Atlantic scombrids. Unpublished manuscript, 9 p. 

 Southwest Fisheries Center, National Marine Fisheries 

 Service, NOAA, La Jolla, CA 92037. 



blad (1966) and modified by Tomlinson (1970)' 

 was used to separate the age classes and estimate 

 the mean lengths. For samples with only one 

 prominent mode, the modal length, or midpoint 

 of length interval of maximum frequency was 

 considered the "mean length." Representative 

 length-frequency distributions are shown in Fig- 

 ure 2. 



Mean lengths for each sample are plotted in 

 Figures 3-6. For each region a serial succession 

 of increasing mean lengths with time was desig- 

 nated a year class, with only one recruited per 

 year although two groups appear to be recruited 



^ Tomlinson, P. K. 1970. Program for separating 

 mixture of normal distributions. Unpublished manu- 

 script, 2 p. California Department of Fish and Game, 

 Operations Research Branch, Long Beach, CA 90802. 



JULT 

 N.309 



AUGUST 

 N-(55 



OCTOBER 

 N=76 



NOVEMBER 

 N=224 



DECEMBER 

 N=592 



80 lOO 120 140 160 160 200 20 40 60 80 100 IZO I40 160 IBO 200 

 FORK LENGTH (cm) 



Figure 2. — Length-frequency distributions of samples 

 from Pointe-Noire, 1970. Arrows indicate mean lengths 

 of modal groups that were identified by curve fitting. 



Figure 3. — Mean length of size groups of yellowfin tuna 

 as a function of sampling date at Abidjan. Growth of 

 the 1963-69 year classes are indicated 



177 



