The ANOVAs showed an effect of the maternal photoperiod and the strain × photoperiod interaction on egg width and egg volume, as opposed to egg length that is not influenced by the explanatory variables ( Table 1). A strain effect on egg

size was only revealed by the MANOVA and not by the ANOVA, underlining that this effect is probably small. The maternal photoperiod is also the factor having the main effect on egg size, even if other KU-60019 price factors may still interact. The mean volume of SD eggs is larger than LD eggs, regardless of the temperate or tropical origin of strains (Kruskal–Wallis = 43.5, df = 1, p < 0.01) ( Fig. 3C). The egg length is not linearly related to egg volume (R2 = 0.31) and represents a relatively small contribution in the calculation formula of the egg volume, as opposed to egg width which is positively correlated with the egg volume (R2 = 0.87). The parameters values used to model reaction norms are presented in appendix (Table A.1). The reaction norms for the tropical strain reared under different day lengths remained closed for each studied trait while reaction norms for the temperate strain reared under SD and LD conditions moved away from each other during embryogenesis (Fig. 4). The difference in the timing of the serosal cuticle, segmentation, ocelli and egg burster appearance selleck kinase inhibitor is described below as measured when 50% of the analyzed

embryos had acquired these traits (Table 2). The LD temperate embryos’ serosal cuticle appeared 5 h later triclocarban than in the tropical strain ones, and in temperate embryos reared under diapause-inducing conditions it appeared almost 10 h later.

This difference between SD and LD temperate eggs was supported by direct observations: embryos aged from 18 to 47 HAE showed different stages of development between diapause-induced and non-diapause-induced embryos (Fig. A.2). The abdominal segmentation appears on the third day of embryonic development, first on LD embryos of the temperate strain. In this strain, diapause-programmed embryos are segmented 19 h after those reared under non-diapause-inducing photoperiod. The segmentation of SD and LD tropical embryos presents a temporal development intermediate to temperate embryos’ development. Pigmented ocelli first appeared at 93 HAE in temperate strains and at 102 HAE in the tropical strain. The ocelli development began at the same time for LD and SD temperate strains, but finished 2 days earlier in the LD temperate strain. Egg burster formation in SD temperate strains took 38 h longer than in LD strains. The maternal photoperiod has a significant effect on the embryogenesis time for all the studied traits in the temperate strain and estimates of c1, representing the distance between turning points, all being positive for SD model ( Table 3).