Best-fitting model of chromosome number evolution, basic chromosome number (x) at the root of the tree with its probability (in brackets), and inferred number of chromosome gains, losses, polyploidy (PP), and demipolyploidy (Demi-PP) events for each dataset.

Significant P values from the Chi-square analyses for the present to 10% of total time (“recent” times) vs. rest of the chronogram tree (“ancient” times) temporal level are shown, comparing the number of observed (inferred by ChromEvol analysis) vs. expected (under the null hypothesis of constant transition rate through time) polyploidy and gains/losses transitions. The type of gains and losses of single chromosomes has been inferred from the bibliography (Supporting Information S1 Notes S2) and from the results of the analysis of the relationship between chromosome number and genome size (Supporting Information S2).
CR =  Constant_Rate, three parameters: gains and losses of single chromosomes and polyploidy; CRD =  Constant_Rate_Demi, three parameters: gains and losses of single chromosomes and one for polyploidy and demipolyploidy; CRDE =  Constant_Rate_Demi_Est, four parameters: gains and losses of single chromosomes, polyploidy and demipolyploidy; CRND =  Contant_Rate_No_Dupli, two parameters: gains and losses of single chromosomes; LR =  Linear_Rate, five parameters: gains and losses of single chromosomes, gains and losses of single chromosomes proportional to chromosome number, and polyploidy; LRD =  Linear_Rate_Demi, five parameters: gains and losses of single chromosomes, gains and losses of single chromosomes proportional to chromosome number, and one for polyploidy and demipolyploidy; LRDE =  Linear_Rate_Demi_Est, six parameters: gains and losses of single chromosomes, gains and losses of single chromosomes proportional to chromosome number, polyploidy and demipolyploidy; and LRND =  Linear_Rate_No_Dupli, four parameters: gains and losses of single chromosomes and gains and losses of single chromosomes proportional to chromosome number.
Alternatives analyses (internal vs. external branches of the chronogram tree, present to 25% of total time vs. rest of the chronogram tree, and present to 50% of total time vs. rest of the chronogram tree) reveal identical conclusions.