46 
Fishery Bulletin 95 ( I ), 1997 
length keys involves the fewest assumptions. Where 
almost certain knowledge of growth and year-class 
strengths is lacking, and the method for ageing the 
fish is robust, this method is probably the best choice 
for monitoring the age composition of a catch. 
Where time-series data for year-class strengths are 
available, where growth and natural mortality are 
reasonably known, and where there are insufficient 
age determinations to construct age-length keys, the 
probabilistic method is clearly superior to age assign- 
ments from inverted growth models and also might 
be as good as, or better than, the age-length keys if 
they were available. Where growth and year-class 
strengths are well characterized and natural mor- 
tality is reasonably known, the probabilistic method 
should outperform all the other alternatives. Addi- 
tionally, this method should be very useful for esti- 
mating the age composition of catches for the most 
recent year of a time series for which sample-age 
analysis may not yet be complete. It should also pro- 
vide a reliable method to estimate the age composi- 
tions of catches for intermediate years of a time se- 
ries, for which insufficient age determinations are 
available to construct age-length keys. 
Acknowledgments 
I thank M. Schirripa, C. Porch, and two anonymous 
reviewers for helpful comments on the manuscript. 
Literature cited 
Beamish, R. J., and D. A. Fournier. 
1981 . A method for comparing the precision of a set of age 
determinations. Can. J. Fish. Aquat. Sci. 38:982-983. 
Goodyear, C. P. 
1989 . LSIM-A length-based fish population simulation model. 
U.S. Dep. Commer, NOAATech. Memo. NMFS-SEFC-219. 
Hoenig, J. M., and D. M. Heisey. 
1987 . Use of log-linear model with the EM algorithm to 
correct estimates of stock composition and to convert length 
to age. Trans. Am. Fish. Soc. 116:232-243. 
Ketchen, K. S. 
1950 . Stratified subsampling for determining age- 
distributions. Trans. Am. Fish. Soc. 79:205-212. 
Powers, J. E., and V. R. Restrepo. 
1992 . Additional options for age-sequenced analysis. Int. 
Comm. Conserv. Atl. Tunas Coll. Vol. Scientific Papers 
39(21:540-553. 
Westrheim, S. J., and W. E. Ricker. 
1978 . Bias in using an age-length key to estimate age-fre- 
quency distributions. J. Fish. Res. Board Can. 35:184—189. 
