16 



Fishery Bulletin 101(1) 



3000 

 2500 

 2000 

 1500 

 1000 

 500 

 



Sep. 7-9 1990 



49/52 stations 



n=5,531 Z=0-37 



r^=0.97 SE=0.05 



2 1 2.6 3,1 3 6 4 1 4.6 5,1 5 6 



700 

 600 

 500 

 400 

 300 

 200 

 100 

 



11 16 2.1 2.6 3.1 3.6 4.1 4.6 5.1 5.6 



Sep. 14-16 1991 



41/50 Stations 



n=3,611 Z=0.20 



r^=0.82 SE=0.03 



1400 

 1200 

 1000 

 800 

 600 

 400 



f 

 200 J J 



Sep. 9-11 1993 



33/47 stations 



n =7,282 Z=0.30 



1^=0.93 SE=0.04 



-fi^JZL 



1 1 1 6 2 1 26 3 1 3.6 4,1 4,6 5.1 5 6 



Sep. 14-16 1990 



49/54 stations 



n=9,948 Z=0.30 



r=0.98 SE=0.02 



^ wn Wl m^ _ 



3 1 36 4 1 46 5 1 5 6 



Sep. 21-23 1991 



45/51 stations 



n =4,379 Z=0.28 



r^=0.97 SE=0.02 



B H PI PI PI - 



1.1 16 2.1 2,6 3,1 3.6 4.1 4.6 5 1 56 



Sep. 19-21 1993 



25/32 stations 

 n = 1,490 Z=0.32 

 r^=0.90 SE=0.05 



11 16 2 1 2 6 3.1 36 4 1 46 5 1 5 6 



Size class (mm) 



Figure 6 



Size-frequency distributions of Atlantic bumper larvae collected during six cruises 

 in the northcentral Gulf of Mexico during September 1990. 1991, and 1993. Values 

 of Z, SE, and r'^ refer to mortality curves produced from the duration-corrected, 

 age-frequency distributions (omitting the smallest size class). The fraction listed for 

 each cruise refers to positive (larvae were collected) stations/total stations sampled. 

 Abundance of each size class is pooled estimates of station abundances. 



that temperatures varied by less than 2°C. Significant dif- 

 ferences in both age versus length and age versus weight 

 relationships were also found for vermilion snapper larvae 

 collected at relatively close stations (i.e. 17 km apart). Wa- 

 ter temperatures at these two stations were similar but 

 differed by as much as 2°C. Faster growing larvae reached 

 a length of 5 mm approximately 2 days sooner than larvae 

 growing in nearby areas. Significant differences were also 

 found in larval weight-gain; dry weight of 11-d-old larvae 

 from different stations varyicd by as much as 657f . 



The variability in growth rates that we observed was 

 likely caused by station differences in food availability 

 and size-selective mortality, and to a lesser degree by wa- 

 ter temperature. Unfortunately our data did not allow 

 us to determine the individual effects of these factors on 

 observed growth rates. At least for Atlantic bumper, the 

 effects of temperature changes were probably minimal. 



Larval survival is generally more influenced by factors 

 other than temperature. Morse (19891 found a positive cor- 



relation between length-dependent mortality and surface 

 water temperature for 26 larval fish taxa and attributed 

 this to increased predator consumption rates (caused by 

 increased metabolic rates) at higher temperatures. He 

 also concluded that increased growth due to increases in 

 temperature alone would generally impart no advantage 

 to reduce larval mortality because of the concomitant 

 increased predatory consumption rates. Increased larv-al- 

 stage duration at cooler temperatures is not necessarily 

 associated with increased cumulative larval mortality be- 

 cause predation rates decline with decreasing temperature 

 (Pepin, 1991; FVancis, 1994). Methot ( 1981 ) concluded that 

 after correcting for the effect of temperature on growth 

 rates, the mean growth rate of larval fish is an indicator 

 of the degree to which lar\'al growth, and presumably sur- 

 vival, is food limited. 



We acknowledge that size-selective predation, i.e. "cull- 

 ing out" the slowest (or fastest! growing lai-vae, could have 

 produced the differences in size-at-age structure among 



