GENETIC AND PHYSIOLOGICAL FLEXIBILITY 455 



not succumbing was arbitrarily set at 50 rather than 60 since 

 exposure to the cold temperature was for only 10 min. 



Special environmental variance (a|j was considered to be a 

 measure of individual flexibility since it is a measure of the range of 

 tolerance within individuals, especially when individuals are exposed 

 to a change in external temperature and the change in their average 

 tolerance (acclimation) can be measured. 



The measurements of additive genetic variance are all based on 

 known genetic relationships, the principle being that any corre- 

 sponding phenotypic correlation (excluding a common environment) 

 is caused by the common genes. Hence similarities and differences 

 among individuals are in part genetic, and the characteristic 

 concerned varies genetically. In this study genetic variance was 

 measured in three ways (Appendix A) by use of the relationships 

 between parents and offspring, between half-siblings, and between 

 full -siblings. As can be seen, the estimate of heritability obtained 

 from the ratio of 2 COVpg to ap is inflated if any nonadditive 

 (dominant and epistatic) variances or common environmental effects 

 are present. The standard errors for heritability estimates were 

 calculated according to Falconer (1963). 



Physiological flexibility can be measured as random variation in 

 temperature tolerance within individuals, directly from test— retest 

 correlations and indirectly by subtracting genetic variance from total 

 phenotypic variance (Appendix B). In the context of adaptation to 

 thermal stress, a more appropriate measure is the variance in 

 tolerance between sets of animals exposed to different temperatures 

 before testing. The details of estimating this variance from the 

 analysis of variance are also shown in Appendix B. The variance 

 between exposure temperatures can also be expressed as a fraction of 

 the total variance. Because the variance between exposure tempera- 

 tures and the total variance are both estimates, in one case the 

 former is actually larger than the latter, as can be seen from the data 

 in Table 3. In addition to its expression as a variance, physiological 

 flexibility can also be shown as the difference in average tolerances 

 (Figs. 1 and 2) or the percentage change in tolerance when two 

 random groups of copepods are exposed to different temperatures 

 before testing for temperature tolerance. 



The range of temperatures over which reproduction occurred was 

 tested by exposing hatched eggs to 4°C and raising temperature 

 gradually to 27 and 29° C, when reproduction was observed at the 

 adult stage. Ten flasks, each containing 20 ovigerous females, were 

 held at 4°C until the egg sacs were dropped and then at 10°C for 2 

 days and 20°C for 2 days. They were then placed in a water bath in 



