TABLE 2 The relationship of the change in skull height in juvenile (July) and adult S. araneus with weather variables between 1953 and 2004. All regressions, except mean daily temperature for juveniles, were corrected for autocorrelation (see Methods).r2 is the proportion of the data explained by each weather variable; significant P values are shown in bold. The direction of significant relationships is shown in parentheses. There were fewer than 52 analysed years because we did not have skull measurements from all years