Nicotra Fan Selection Software Download
to assess whether there is a tendency for more significant selection coefficients on the traits quantified in the selection gradient studies to involve greater plasticity, we found the original study on which each selection coefficient was based (figures 1 and 2 ) and used the midpoint between the minimum and maximum trait values in the study to calculate a plasticity score for each trait in each study (high values indicating greater plasticity). this revealed that in almost all cases (n=25) the original study had quantified the trait at a larger value for the low-plasticity than high-plasticity group. of the 20 examples of positive selection coefficients, 18 had significant higher mean plasticity scores in the high-plasticity group than the low-plasticity group, including all three significant selection coefficients. the single example of significant negative selection was for mean plasticity in arabidopsis flowering time [ 68 ] and there, the mean plasticity score in the high-plasticity group was less than half of that in the low-plasticity group. two of the seven examples of stabilizing selection had higher mean plasticity scores in the low-plasticity group than in the high-plasticity group. hence, in almost all cases, studies that found evidence for selection on plasticity also found that the trait for which selection was estimated was measured at a higher value for the low-plasticity group. our two studies of selection on phenotypic plasticity in the context of rapid climate change [ 38, 68 ] are in some ways in the same experimental vein as the previous review by van buskirk & steiner [ 55 ], but we have focused on a different set of data. the latter had searched all published studies, whereas we only searched studies of thermal environments and had restricted our literature search to the last 5 years, which we feel has allowed a much better chance of detecting evidence of plasticity and selection. however, we also note that our sample size of studies is small relative to that of van buskirk & steiner [ 55 ], who had systematically reviewed studies over the previous 30 years and had identified a total of 79 estimates of selection on plasticity. nicotra fan selection software download we next sought to examine whether the directions of selection coefficient estimates within and across studies was consistent. we found that the directions of linear selection coefficients were consistent across studies in more than half of the cases (figure 3 ). there were five cases of contradictory directions, with contrasting directions of selection between studies of the same phenotypic trait and the same species, in a. thaliana flowering time [ 68 ], height in wahlenbergia [ 38 ], size in prunus [ 86 ], and breeding time in o. cuniculus [ 83 ]. two of these contradictory cases are of the same studies as those that found evidence for a significant selection coefficient. there were also two studies of larus argentatus [ 84 ] and p. bairdii [ 85 ] that found evidence for stabilizing selection on plasticity. in one case (that of a. thaliana flowering time [ 68 ]), the study was based on a single, naturally-occurring population and thus was not a controlled experiment. in another case (that of p. bairdii breeding time [ 83 ]), there were two environmental conditions but only one condition of interest and so the study was not controlled. d8a7b2ff72
the potential use of the plasticity-fitness correlations to infer the nature of selection on plasticity was not, however, supported by the cross-study meta-analysis. the three meta-analyses that we conducted did not detect significant correlations between the plasticity slope and individual fitness (meta-analysis one, r2 = 0.086, p = 0.434; meta-analysis two, r2 = 0.091, p = 0.386; meta-analysis three, r2 = 0.116, p = 0.259).
overall, the results of the eight independent studies strongly suggest that in many cases there is no selection on plasticity, and that selection on plasticity is not consistent, but rather is dependent on the scale of the environmental gradient. this is not to say that the relative importance of environmental variance in determining the nature of selection on plasticity is negligible, but rather that the evidence to date suggests that the most important factor is the magnitude and nature of the environmental gradient. there are several possible explanations for the inconsistent nature of the evidence for selection on plasticity, including spatial and temporal variation in selection and genetic variance-covariance matrices, which may be important but are not addressed here.
a key conclusion of our systematic review is that selection on reaction norms is only a modest contributor to overall plasticityfitness gradients in the literature so far. this finding is supported by the cross-study meta-analyses of plasticity-fitness relationships, which did not detect any significant relationships between the plasticity slope and fitness. given the widespread plasticityfitness relationships, this in turn provides evidence for a large number of studies not identifying any evidence of selection on plasticity at all.