Table 1.
List of extinct archosaur taxa/ootaxa with estimated egg mass (M) and eggshell porosity (Ap∙Ls-1) used in this study.
Fig 1.
Porosity of archosaur eggshell.
Schematic diagram of archosaur eggshell with high porosity (A) and low porosity (B), modified from [18]; tangential thin sections of living covered nester Caiman latirostris (C), living open nester Pavo cristatus (D), and non-avian maniraptoran Troodon formosus (E). Abbreviations; A, individual pore area; D, pore density; Ls, pore length. Arrows indicate pore canals.
Table 2.
List of variables used for this study, modified from Tanaka and Zelenitsky [18].
Table 3.
List of equations used for this study, modified from Tanaka and Zelenitsky [18].
Fig 2.
Bivariate plot of eggshell porosity and egg mass between living covered and open nesters.
Eggshell porosity relative to egg mass is highly correlated to nest types (p < 0.01), as reflected by the different regression models between closed and open nesters.
Table 4.
Results of conventional OLS regression models for living archosaur species.
Table 5.
Results of conventional and phylogenetically-corrected ANCOVA for living archosaur species.
Fig 3.
Bivariate plot of eggshell porosity and egg mass in both living and extinct archosaur taxa/ootaxa.
Titanosaurs and Lourinhanosaurus show high eggshell porosity, comparable to living species with covered nests. In contrast, oviraptorosaurs, Troodon, and moas show lower eggshell porosity, similar to species with open nests.
Table 6.
Cross-classification/ confusion matrix from LDA and pFDA.
Table 7.
Inferred nest types for extinct archosaurs based on the linear discriminant analysis.
Fig 4.
Misclassification rate of pFDA for living species through changing Pagel's lambda values.
Dash line shows the optimal lambda value of 0.56. Note that the overall misclassification rate increases with increasing lambda values from 0 to 1.
Fig 5.
Comparison of the discriminant function between covered and open nesters in living and fossil archosaurs.
Horizontal bars inside boxes represent medians, lower and upper ends of boxes are the 25% and 75% quartiles, respectively, and whiskers represent the smallest and largest cases. Outliers are represented by dots and extremes by diamonds. Note that covered nesters show relatively lower values than open nesters.
Fig 6.
Inferred nest type for six extinct archosaurs as a function of Pagel's lambda values.
Inferred nest type is generally consistent across all lambda values, except for oviraptorosaurs where inferred nest type changes when the lambda value varies between 0.08 and 0.52, and for titanosaurs and Lourinhanosaurus, which change to open nesters when lambda values approach one. The yellow line indicates the optimal lambda value (0.56).
Table 8.
Inferred dinosaur nest types based on the phylogenetic flexible discriminant analysis.
Fig 7.
Evolution of nest types among archosaurs.
(A) Phylogeny of archosaurs with inferred nest types based on eggshell porosity and taphonomic evidence. Covered nests are the primitive condition for dinosaurs; open nests and brooding behavior were present among non-avian maniraptoran theropods but may have first appeared earlier. Although the eggs of early open nesters were still partially covered by substrate, open nests with fully exposed eggs likely arose among Euornithes. (B) Phylogeny of Neornithes with inferred nest types based on eggshell porosity (Emeidae) and literature (other birds). Open nests with fully exposed eggs are the primitive condition for modern birds, although secondary reversal to partial egg burial occurred independently in several clades. Information for bird orders which include species that partially bury the eggs (Charadriiformes) or occasionally cover the eggs in open nests (Accipitriformes, Anseriformes, Charadriiformes, Gruiformes, Passeriformes, Podicipediformes, Struthioniformes, Tinamiformes) was taken from [97–103]. Cladograms are based on [73,88,104–106].