The Euler-Lotka equations form the basis of foundational definitions of biological fitness. However, these equations constitute a one-sex model, which fails under simple biological and mathematical constraints. Here, I discuss the effects of these errors on evolutionary theory. In particular, I discuss the contrast between direct and inclusive fitness models on the evolution of human ageing, demonstrating how Hamilton’s proof of declining selection under the Euler-Lotka model is less predictive of human mortality than inclusive fitness models. Finally, I discuss the discovery of huge diversity in human ageing rates, across 39 million US mortality records and 263 years of population data. The potential adaptive plasticity of human ageing rates within these data is explored, with reference to machine learning models. Pointing to the future, I discuss the current expansion of this work to family trees of 94 million people.

Saul Newman received his PhD from the John Curtin School in 2015, on the evolution of life history traits. He is currently a post-doctoral fellow at the CSIRO, applying machine learning to predict life history and reproductive traits in wheat, using large-scale genome and environment data.