HAYLIFF: INTEGRITY OF SKIFMACK TUNA SCHOOLS 



Table ^.—Continued. 



This means that for individual dates for recaptures 

 during June (i.e., during the month of release), 3 of 

 11 tests equivalent to those in Table 4 with 1 degree 

 of freedom were significant at the 5% level; for 

 recaptures during July, 2 of 17 tests equivalent to 

 those in Table 4 with 1 degree of freedom were sig- 

 nificant; and so on. From these tests it appears that 

 after 1 month at liberty the tagged and untagged 

 fish are randomly mixed with one another in some 

 cases, and after 3 to 5 months at liberty they are 

 randomly mixed with one another in nearly all cases. 



r= I 



(T, - F,Pf 

 t=i F,P (1 - P) 



where T, = number of tagged fish in set z, 



Fj = number of tagged and untagged fish in 



set i, 

 n = number of sets, and 

 P = probability of a fish being tagged = 



« n 



I (T,)/I (F,) 



i = l 



! = 1 



Binomial Homogeneity Tests 



For the binomial homogeneity test the formula 



is used. Computer program SCHOOLA was used for 

 this purpose. 

 The results are given in Table 6. In general, the 



637 



