This study tests a long-time established assumption, the existence of a trade-off between maximum size of species and the size at which they can begin reproduction. The authors use a large dataset and a modeling approach to test whether this hypothesis holds for a variety of species from different biomes.

Overall, I think this is a timely and relevant study, built upon a large and solid dataset (but see some concerns about the dataset in my comments below). The hypotheses are very clear and well explained, with a conceptual model that is used to assess several alternative hypotheses about how size at maturation and maximum size are related, and the implications of each one.

I don’t really have any big concerns about this study, and I think it can make an interesting contribution for Ecology Letters and most of its readers. The results are robust and the discussion well grounded, and in general this is a nice contribution to forest ecology. I do have, however, some comments that hopefully will contribute to improve the work:

1) Although Figure 2 is a nice attempt to present a conceptual diagram of the methodology used, I think it would benefit from a bit more work on relating the different steps. For instance, it is not clear enough how predictions of maturation size from the individual model are different from the maturation size derived from the proportional cost model, or if both are related in some way. Although this is clearer in the text, I think it would be helpful to try to represent it in the figure.

2) The MASTIF data, although an impressive dataset covering 12 million observations, 898 species and 112 families, seems to be biased towards North American and European species. Indeed, although most species (54%) come from South America, 97% of plots are in North America or Europe. That means that data series for South American species are necessarily shorter, or less complete. In this regard, I miss some information on how this could affect the predictions of the model. Some discussion in this regard would be very beneficial for the study.

3) Related to the previous point, the phylogeny by Zanne et al. (2014) contains 400 species, i.e. around 50% of the species included in your dataset. Did you check if it somehow reinforced the bias I mentioned above? I mean, are species in the phylogeny predominantly in some biomes?

4) Derivation of tree maturation size from MASTIF model: although this section derives from the work by Clark et al. (2019), I think some steps are not explained in sufficient detail. For instance, I believe equations 3, 4 and 5 would require more detailed explanations for most readers to understand what is being done.