highly significantP < 0.0002
Contribution to the Next Years’ Spring Peak We investigated the correlation between the abundance of the spring generation (dependent variable) and the abundance of either of the two generations the year before (independent variables) with a multiple linear regression model: the contribution of the second generation to the abundance in spring next year is highly significant ( R 2 = 0.80, P < 0.0002), while the contribution of the first generation is not significant ( R 2 = 0.17, P = 0.21) ( Figure A8 a).