Gaughan et al.: Distribution of Sardinops sagax off southwestern Australia 



625 



West Coast 



114.5 115 1155 116 





1996 



1998 



... 



1999 



1145 115 1155 



Figure 2 (continued) 



arguments, a, must be zero; therefore the above family 

 of models reduces to 



(B spdepm\ = a (PPS,)^ + e, 



(2) 



An attempt to fit the linear regression model, composed 

 of the natural logarithms of either side of the above rela- 



tion, to the observed data was unsuccessful because of 

 heterogeneity of error variance. 



Direct estimation of model II with a nonlinear regres- 

 sion procedure gave estimates of A that were near 1 and 

 with large standard errors on account of the small size 

 of the data sets. However, residual diagnostics were sat- 

 isfactory. Because the observed PPS values fell between 



