316 



Fishery Bulletin 101(2) 



1998/99 WDS 



1999/2000 WDS 



I St mode 



B 



Isi mode 



nd mode 



3rd mode 



4th nutde 



96 16 



Lipofuscin index 



032 0.48 0.64 0.8 0.96 



Figure 3 



Frequency histograms of lipofuscin index (A and B) in eyestalks of crabs collected in 1998-99 (f!=317) and 1999-2000 

 (.11=259). Solid lines are derived from a predicted frequency distribution and dashed lines show normally distributed com- 

 ponents of each mode. Age assignments of each mode in lipofuscin index histograms and size classification are shown in 

 Table 3 and Figure 4. 



abundance with the Hpofuscin index, consistent with a pre- 

 dominate pattern of mortality among age classes. Based on 

 lipofuscin analysis, the relative abundance between age 

 classes (age 1 and 2+ classes) was not significantly different 

 between two sampling years (?-test; P>0.05). 



Discussion 



Size-specific patterns of crab abundance in the Chesapeake 

 Bay exhibit significant interannual variation; a signifi- 

 cant decline in juveniles was seen during the second year 

 (1999-2000) of the study (Fig. 1, A and B). Such variations 

 have been observed previously in Chesapeake Bay (Abbe, 

 1983; Hines et al., 1987; Lipcius and Van Engel, 1990), and 

 may result from interannual variation in recruitment of 

 larvae, postlarvae, and juveniles (related to the spawning 

 stock), physicochemical conditions, and food availability 

 (Holland et al., 1987; McConaugha, 1988). Density-depen- 

 dent processes such as cannibalism and predation may 

 also regulate juvenile abundance and dampen recruitment 

 variation, particularly when stock levels are high (Lipcius 

 et al., 1995; Kahn et al., 1998). 



Age estimates based upon lipofuscin index were more 

 highly resolved than those based on CVV, and there were 

 at least three age modes (ages 0, 1, and 2) apparent in 

 both sampling years. Perhaps more importantly, individu- 

 als were classified differently with lipofuscin than with 

 (;W, which indicated that age estimates based upon CW 

 could lead to substantial errors in determination of growth, 

 mortality, and fishery yield estimates. Recent growth-rate 

 measures have indicated that ('hesapeake Bay blue crabs 

 may grow substantially faster than previously recognized, 

 but also that some crabs may grow at very slow rates under 



unfavorable environmental conditions (Ju et al., 2001). In 

 other words, lipofuscin-based age determinations indicate 

 highly variable individual growth rates. We speculate that 

 such variability is the result of the discontinuous nature 

 of crab growth combined with the protracted spawning 

 season and strong seasonal pattern in growth (driven 

 by temperature), which together amplify relatively small 

 differences in hatching dates and early growth rates into 

 large differences in carapace sizes. Spatial heterogeneity 

 in habitat would also be a contributor to high variability 

 in crab growth rates. Alternatively, the degree with which 

 lipofuscin refiects true chronological age may have contrib- 

 uted to the high variation seen in CW-based ages versus 



