HOUDE: VITAL RATES. AND ENERGETICS OF MARINE FISH LARVAE 



tures (Fig. 2), indicating more scope for stage 

 duration at low temperatures and presumably 

 in high latitudes. This result was opposite to 

 that observed for gi'owth rate, in which there 

 was more variability at high temperatures (Fig. 

 1). 



Mortality 



Mortality coefficients of feeding-stage larvae 

 reported for 22 species ranged from Z = 0.01 to 

 0.69 d"', equivalent to 1-50% d~' mortal- 

 ity rates (Table 4; Fig. 3). Predicted mortality 

 rate (Z) increased approximately 0.01 per 

 degree increase in temperature (7^. 



Z = 0.0256 -t- 0.0123 T 



r- = 0.41 Sh = 0.0034. (3) 



Estimated Z increased from 0.09 at 5°C to 0.40 

 at 30°C. 



Although quite variable among species, the 

 relationship between mortality rate and temper- 

 ature is significant {P < 0.01). Only Atlantic 

 mackerel appeared to be an outlier from the 

 regression line (Fig. 3); its observed mortality 

 rate was higher than expected for species in the 

 15-20°C range. 



Relationship Between Mortality and 

 Grovi'th Rates 



The derived relationship between mortality 

 rate {Z) and growth rate (G) for marine fish 

 larvae was 



Z = 0.0303 + 1.3085 G. 



(4) 



Also, from predicted G and Z in Equations (1) 

 and (3), it is apparent that the ratio GIZ in- 

 creases at high T. 



TCC) G Z G/Z 



5 0.043 0.087 0.494 

 15 0.137 0.210 0.652 

 30 0.278 0.395 0.704 



Despite the elevated mortality rates suffered by 

 marine fish larvae at high temperatures, their 

 growth potential may allow such larval popula- 

 tions to accumulate biomass and to survive at 

 relatively high rates when feeding conditions are 

 favorable. 



Net Survivorship 



Predicted net survival at metamorphosis in- 

 creased by an approximate factor of nine as tem- 

 perature increased from 10° to 30°C. 



The effects on TV^^ of a 10% dechne in G, a 10% 

 increase in Z, or a combination of 10% dechne in 

 G and 10% increase in Z were large. 



100 .V^et C^c) 0.00005 0.0003 0.0007 



Effects on percent larval survival at metamor- 

 phosis (100 A^niet) of declining growth rates or 

 increasing mortality rates are gi'eater at low 

 than at high temperatures, a consequence of the 

 long stage durations at low temperature and 

 their dependency on growth rate. Calculated 

 ratios of A^,,,,.; from the examples given above, 

 termed relative survival, compared to those 

 expected from the regression relationships at 

 average G, D and Z, are 



479 



