Stock (ancestral) cultures of Selenastrum capricornutum grew optimally in low Cu liquid nutrient medium (LNM)containing 0.004 pg L-' Cu2+. Growth of stock cultures was optimized on low Cu solidified nutrient medium (SNM) containing 20 pg L-l Cu2+. Cell strains (clones) were isolated from unusually large colonies on high Cu SNM (1,000 pg L-' Cu2+). Most isolated strains performed the same as ancestral populations in high Cu LNM (50 pg L-' Cu2+), but three strains grew more slowly. These "variant strains" were Cu tolerant on SNM (equal growth with low and high Cu) and produced larger colonies than did the ancestral population on both low and high Cu SNM. In contrast, colonies from ancestral cell samples grew more slowly in high versus low Cu SNM. No significant growth differences were observed between ancestral populations and variant strains in low Cu LNM, suggesting low fixed costs of adaptation to SNM. However, slower growth of variant strains in high Cu LNM constituted a fitness trade-off associated with adaptation. Rapid adaptation to pollution may often result in important biological trade-offs even when costs of adaptive mechanisms are low.

Ipomopsis rubra plants grown in the laboratory initially produced hermaphrodite flowers, but some self- or sib-mated individuals switched to produce large numbers of pistillate (male sterile) flowers. The sex change did not occur with outcrossing. Plants with extreme male sterility were also observed in natural populations, usually in smaller individuals. Male sterility may be compensated by more seeds (resource reallocation), better seeds (avoidance of selfing), or both. Pistillate flowers were smaller, so savings could be used for additional seeds. Selfed seeds had reduced survival and fecundity, so avoidance of selfing could produce better quality offspring. We explored costs and benefits of sex change with two fitness models. The first assumes randomoutcross matings. Estimates of resource reallocation and inbreeding (selfing) depression are sufficient for pistillate inflorescences to have equal or greater fitness than hermaphrodite inflorescences if the selfing rate is high. Frequencies of sex change with intensive self-pollination were consistent with this model. The second model assumes all nonself matings are between sibs in "local mating" groups. Parents may benefit by male sterility in offspring, but gains would be higher if sex change occurred earlier and at higher than observed frequencies.

Abstract-A simple, manually diluted, semicontinuous, 96-h algal growth assay technique was developed to measure changing rates of population growth following sublethal chemical exposure. Rates were estimated directly from changing cell counts in a fixed volume of cell suspension. Short-term rate fluctuations in exponential rate parameters were observed in Selenastrum capricornutum populations using this method and similar fluctuations were also documented by reanalyzing conventional static culture assay data. Replicate cultures tended to fluctuate in unison, and patterns of population increase were similar in static assays initiated on different dates. The latter suggested that nonuniform rates of S. capricornutum population increase were not due simply to environmental variation. All populations were preacclimated to test conditions, but growth lags were consistently observed for 12-24 h following inoculation. Subsequent rate fluctuations probably resulted from a high degree of cell-cycle synchronization. Treating systematic rate fluctuations as random error lowered measurement precision, especially with respect to estimates of rate changes over time. Systematic variance may be difficult to eliminate in practice, but repeated-measures regression methods can account for this effect and substantially reduce rate parameter confidence intervals. Findings are expected to apply to endpoints such as dry weight, total cell volume, chlorophyll, or DNA.

Ipomopsis rubra plants grown in the laboratory initially produced hermaphrodite flowers, but some self‐ or sib‐mated individuals switched to produce large numbers of pistillate (male sterile) flowers. The sex change did not occur with outcrossing. Plants with extreme male sterility were also observed in natural populations, usually in smaller individuals. Male sterility may be compensated by more seeds (resource reallocation), better seeds (avoidance of selfing), or both. Pistillate flowers were smaller, so savings could be used for additional seeds. Selfed seeds had reduced survival and fecundity, so avoidance of selfing could produce better quality offspring. We explored costs and benefits of sex change with two fitness models. The first assumes randomoutcross matings. Estimates of resource reallocation and inbreeding (selfing) depression are sufficient for pistillate inflorescences to have equal or greater fitness than hermaphrodite inflorescences if the selfing rate is high. Frequencies of sex change with intensive self‐pollination were consistent with this model. The second model assumes all nonself matings are between sibs in “local mating” groups. Parents may benefit by male sterility in offspring, but gains would be higher if sex change occurred earlier and at higher than observed frequencies.

Stock (ancestral) cultures of Selenastrum capricornutum grew optimally in low Cu liquid nutrient medium (LNM) containing 0.004 μg L−1 Cu2+. Growth of stock cultures was optimized on low Cu solidified nutrient medium (SNM) containing 20 μg L−1 Cu2+. Cell strains (clones) were isolated from unusually large colonies on high Cu SNM (1,000 μg L−1 Cu2+). Most isolated strains performed the same as ancestral populations in high Cu LNM (50 μg L−1 Cu2+), but three strains grew more slowly. These “variant strains” were Cu tolerant on SNM (equal growth with low and high Cu) and produced larger colonies than did the ancestral population on both low and high Cu SNM. In contrast, colonies from ancestral cell samples grew more slowly in high versus low Cu SNM. No significant growth differences were observed between ancestral populations and variant strains in low Cu LNM, suggesting low fixed costs of adaptation to SNM. However, slower growth of variant strains in high Cu LNM constituted a fitness trade‐off associated with adaptation. Rapid adaptation to pollution may often result in important biological trade‐offs even when costs of adaptive mechanisms are low.