242 



Fishery Bulletin 94(2), 1996 



(36.6 m), suggesting that sea turtle stranding rates 

 increased as fishing intensities increased in these 

 depth intervals. The null hypothesis of homogeneity 

 among the ten correlations was not rejected (X 2 =9.5, 

 P >0.05 ), showing that the correlations did not differ 

 significantly from one another. Thus, the null hypoth- 

 esis that the statistical relationship between monthly 

 sea turtle stranding rates and shrimp fishing inten- 

 sities in the northwestern Gulf was the same in 1990- 

 93 as in 1986-89 could not be rejected. 



Four significant negative correlations between 

 ln(S + 1) and ln(£ + 1) were also detected for depth 

 intervals seaward of 20 fm. Two occurred on the up- 

 per coast, one in 1986-89 for \n(E + 1) in >45 fm 

 (r=-0.358, P=0.0125), and one in 1990-93 for 



ln(£ + 1) in the 35-40 fm depth interval (r=-0.334, 

 P=0.0203 ). Two occurred on the lower coast in 1986- 

 89, one for ln(£ + 1 ) in the 20-25 fm depth interval 

 (r=-0.431, P=0.0022), and the other in the 25-30 fm 

 depth interval <r=-0.308, P=0.0332). 



A temporal-spatial pattern in monthly fishing in- 

 tensities was evident from the correlations involv- 

 ing the ln(£ + 1) of the ten depth intervals (Tables 

 5-8). Significant correlations between the ln(£ + 1) 

 of adjacent depth intervals tended to be positive, 

 showing that monthly fishing intensities in such 

 depth intervals increased or decreased in synchrony. 

 Conversely, monthly fishing intensities in widely 

 separated depth intervals varied in opposite direc- 

 tions, as indicated by significant negative correla- 



