GKEENBLATT ASSOCIATIONS OF TUNA WITH FLOTSAM 



SKIPJACK SCHOOLFISH 



SKIPJACK ASSOCIATED 

 WITH FLOTSAM 



30 *0 50 60 'G 80 90 100 ilO l£0 tJO 30 40 50 60 '0 80 9C' lOO mO 



20r 



Figure 3— Length-frequency distribu- 

 tions of yellowfin and skipjack tuna 

 caught in unassociated schoolfish and 

 flotsam-associated sets Data collected 

 1973-75 in the CYRA (see Table 2). 



YELLOWFIN SCHOOLFISH 



I I I I I I 'I 



FORK LENGTH (cm) 



YELLOWFIN ASSOCIATED 

 WITH FLOTSAM 



20 1- 



TABLE 4. — Average yearly rainfall (centimeters) in Central 

 America in 1963-75, (Source; U.S. Department of Commerce.) 



1963 64 65 66 67 68 69 70 71 72 73 74 75 

 YEAR 



Figure 4. — Percentage of total unregulated sets that were as- 

 sociated with flotsam in the CYRA. 1963-75. 



small similarities, indicating that the supply of 

 suitable flotsam was not greatly influenced by 

 rainfall. 



Comparing the average catch per set of different 

 set types indicates the relative importance of each 

 set type to the fishery as well as showing trends in 

 the catch rate (Figure 6). All set types had similar 

 catch rates in 1963-66. Porpoise sets and flotsam- 

 associated sets gave much higher catch per set 

 than unassociated sets in 1971-75. One sees that 

 flotsam-associated sets have been the most valu- 

 able set type for the tuna fisherman since 1971. 



Fork length data was stratified by set type and 

 species. The mean length, standard deviation, and 

 sample size were calculated on a single set basis 

 (Table 5). The average standard deviation of fork 

 length of yellowfin and skipjack tuna associated 

 with flotsam was larger than the standard devia- 

 tion found in unassociated sets, though the mean 

 fork length of flotsam-associated sets was smaller. 

 The probability of getting the results shown (Ta- 

 ble 5) by chance was calculated using Kruskal- 

 Wallis one-way analysis of variance (Siegel 

 1956: 184) (Table 5 1. The greater variability of fork 

 length of flotsam-associated tuna supports the 

 hypothesis that flotsam aggregates tuna. 



The yellowfin and skipjack tuna catch distribu- 

 tion on flotsam-associated sets was compared with 

 unassociated schoolfish sets. The average catch 

 per successful set was calculated and the data 

 were plotted as histograms of tonnages using an 

 arbitrary interval of 5 tons (Figure 7). The main 



151 



