Research ArticleEVOLUTIONARY BIOLOGY

Shrinking dinosaurs and the evolution of endothermy in birds

See allHide authors and affiliations

Science Advances  01 Jan 2020:
Vol. 6, no. 1, eaaw4486
DOI: 10.1126/sciadv.aaw4486
  • Fig. 1 The evolution of endothermy and miniaturization in the theropod lineage leading to birds.

    (A) The cost-benefit to switch from ectothermy to endothermy for different ranges of body size was quantified with the Scholander-Irving model, which describes how a rise in metabolism at rest (cost) increases the thermal niche TbTa (benefit). Because there is no thermal gradient between the organism and the environment in the absence of heat production, this curve intersects the abscissa at Tb = Ta when MR = 0 (8). The solid blue and red lines depict the metabolic curves of a typical ectotherm and endotherm, respectively, and the open symbols depict the maximal thermal gradient TbTa possible with resting metabolic rates, used in our model (Eq. 2). (B) A reduction in body size, consistent with the one described from ancestral theropods to basal birds (22), constitutes the evolutionary path of least resistance as the energy costs of being large are traded for those of being endothermic.

  • Fig. 2 Reconstruction of metabolic levels and thermal niche of theropods.

    (A) Theropod phylogeny (15) with branches color-coded according to reconstructed metabolic levels. (B) Scaling of metabolic rate versus body mass (12) for ectotherms (MR = 0.68mass0.75) and endotherms (MR = 3.4mass0.75) and the predicted trajectory of the bird stem lineage during the transition from ectothermy to endothermy. Dashed lines show fold differences between ectotherms and endotherms (1× to 5×); open and closed symbols depict reconstructed values for the bird stem lineage and the tips of the phylogeny, respectively (see Methods). (C) Scaling of thermal conductance C and body mass (13) for ectotherms (C = 2.5mass0.5) and endotherms (C = 1.0mass0.5), fold differences from 2.5× to 1×. (D) Thermal gradient and fold differences calculated with Eq. 1 and values in (B) and (C). The log-log linear trajectories connecting MR and C of the ectothermic ancestor and the endothermic descendant, as well as the resulting trajectory in thermal gradient, are shown with the continuous lines.

  • Fig. 3 Body size evolution and the cost-benefit of endothermy.

    (A) The miniaturization from Tetanurae to basal birds inferred from the fossil record (15), contrasted against 100 simulated size trajectories starting from the same ancestral body size for illustrative purposes (note that for the subsequent full null model, the ancestral body size is allowed to vary). The error represents the SD in reconstructed values across 20 candidate trees (see Methods). (B) The frequency distribution of body mass ratios obtained across 10,000 simulated body size trajectories (histogram) and the energy costs to evolve endothermy expressed per degree Celsius (Eq. 2) under this null model (gray symbols). In this case, the ancestral body size was obtained from a uniform distribution ranging between 10 g and 100,000 kg. The empirical estimate in the bird stem lineage is shown in red. The region in which a reduction in body size would compensate for the energy costs of evolving endothermy, enabling the population to increase in a scenario of constant resources, is highlighted in gray. The arrow depicts the expected population fold increase, given the observed body size reduction in the bird stem lineage as endothermy evolved. These analyses indicate that the energy costs to evolve endothermy are reduced with miniaturization and, as a result, population size may have increased despite the metabolic costs of an endothermic lifestyle.

  • Fig. 4 Tempo and mode in the evolution of endothermy.

    (A) Reconstructed temporal course of metabolic evolution in the bird stem lineage, with dashed lines showing how reconstructions change assuming that either Paraves or Neornithes were fully endothermic instead of the basal bird [for calculations with Neornithes, we assumed a body size of 150 g based on estimates for Vegavis (22) and a time estimate of 100 Ma ago (39)]. The fold increase in MR was calculated by dividing the reconstructed MR during the transition to endothermy by the MR expected for a similar-sized ectotherm and is therefore dimensionless and independent of body size. (B) The evolutionary path of least resistance from ectothermy to endothermy includes inertial homeothermy as a transitional stage, followed by an increase in metabolism concomitantly with a reduction in size. (C) Hypothetical sequence of evolutionary transitions in the bird stem lineage, which combines results from this study with phylogenetic reconstructions of epidermal structures (24, 42) and capacity for active flight (38) (see the main text).

Supplementary Materials

  • Supplementary material for this article is available at http://advances.sciencemag.org/cgi/content/full/6/1/eaaw4486/DC1

    Fig. S1. Relationship between body size reconstructions performed by Benson et al. (15, 23) and Lee et al. (22).

    Fig. S2. Comparison between the topologies of the theropod phylogeny reconstructed by Lee et al. (22) and Benson et al. (15, 23).

    Fig. S3. Replicate of Fig. 2, except that, in this case, analyses were replicated using the dataset and phylogeny by Lee et al. (22).

    Fig. S4. Replicate of Fig. 4, except that, in this case, analyses were replicated using the dataset and phylogeny by Lee et al. (22).

    Fig. S5. Comparison between reconstructed metabolic levels along the bird stem lineage using the dataset by Benson et al. (15) and Lee et al. (22), plotted against the 1:1 line.

    Fig. S6. Phenotypic variance simulated with the difference parameters fitted by Benson et al. (15) for the theropod phylogeny (parameters available in their appendix S5).

    Fig. S7. Simulated OU model overlapped against the empirical data from Benson et al. (15) (their appendix S5), which shows that this model can replicate the distribution of phenotypic data observed along the theropod phylogeny and provide a valid “null model” in the absence of directionality (see below).

    Fig. S8. Results from the null model in the main text compared against expectations for a more conservative model assuming Brownian motion.

  • Supplementary Materials

    This PDF file includes:

    • Fig. S1. Relationship between body size reconstructions performed by Benson et al. (15, 23) and Lee et al. (22).
    • Fig. S2. Comparison between the topologies of the theropod phylogeny reconstructed by Lee et al. (22) and Benson et al. (15, 23).
    • Fig. S3. Replicate of Fig. 2, except that, in this case, analyses were replicated using the dataset and phylogeny by Lee et al. (22).
    • Fig. S4. Replicate of Fig. 4, except that, in this case, analyses were replicated using the dataset and phylogeny by Lee et al. (22).
    • Fig. S5. Comparison between reconstructed metabolic levels along the bird stem lineage using the dataset by Benson et al. (15) and Lee et al. (22), plotted against the 1:1 line.
    • Fig. S6. Phenotypic variance simulated with the difference parameters fitted by Benson et al. (15) for the theropod phylogeny (parameters available in their appendix S5).
    • Fig. S7. Simulated OU model overlapped against the empirical data from Benson et al. (15) (their appendix S5), which shows that this model can replicate the distribution of phenotypic data observed along the theropod phylogeny and provide a valid “null model” in the absence of directionality (see below).
    • Fig. S8. Results from the null model in the main text compared against expectations for a more conservative model assuming Brownian motion.

    Download PDF

    Files in this Data Supplement:

Stay Connected to Science Advances

Navigate This Article