Fig. 1

Correspondence between evolutionary conservation and phenotypic variation during embryogenesis in medaka embryos. a Schematic representation of the relationship between developmental stability and evolutionary conservation of the body plan establishment period in vertebrates [4, 5, 7, 10,11,12]. This hypothesizes that phenotypic variation in the absence of genetic diversity (right) is correlated with evolutionary diversity (left). b Wild medaka populations (Kasasa and Oura) used for measuring microevolutionary conservation. Whole embryonic transcriptomes were compared between gender-matched embryos from each population raised in the same environment. c Microevolutionary conservation evaluated in four developmental stages. Variance of distribution of differential gene expression levels was used to represent intraspecies phenotypic differences (See also ‘Methods’). d An inbred medaka strain (Hd-rR) was used to estimate phenotypic variations. Whole embryonic transcriptomes of gender-matched twins raised in the same environment were compared. e Whole embryonic phenotypic variations were quantified by the variance of distribution of differential gene expression levels (See also ‘Methods’). The Kruskal–Wallis test (P value shown) followed by multiple comparisons (Steel–Dwass) indicated that st. 23.5 and st. 28 had significantly smaller phenotypic variation and intraspecies diversity than the earlier and later stages (st. 15 vs. st. 23.5, P = 8.4 × 10^–3; st. 15 vs. st. 28, P = 1.1 × 10^–3; st. 15 vs. Hatch, P = 1.3 × 10^–2; st. 23.5 vs. st. 28, P = 0.91; st. 23.5 vs. Hatch, P = 1.3 × 10⁻², st. 28 vs. Hatch, P = 1.3 × 10⁻²) and significantly smaller phenotypic variation (st. 15 vs. st. 23.5, P = 4.7 × 10⁻²; st. 15 vs. st. 28, P = 1.7 × 10⁻²; st. 15 vs. Hatch, P = 5.7 × 10⁻²; st. 23.5 vs. st. 28, P = 0.79; st. 23.5 vs. Hatch, P = 1.1 × 10⁻²; st. 28 vs. Hatch, P = 5.8 × 10⁻³). Box plots: centre line, median; box limits, upper and lower quartiles; whiskers, 1.5× interquartile range; points, outliers (d, e)