In this year's fourth issue of ATLA, we have an impressively diverse set of papers, including a 50th Anniversary Article that leads us to reconsider a few notions, and several other papers that will be included in our Virtual Special Collection on Biobanking and the Use of Human Tissues. We are also pleased to announce the recipients of the Michael Balls Award 2021. These delighted winners are the authors of the paper in Volume 49 that the Associate Editors and Editorial Board of ATLA identified as potentially making the most significant contribution to the reduction, refinement and/or replacement of animal experimentation.In our 50th Anniversary Article this time, Landsiedel et al. 1 make a thought-provoking point: that the methods and data accepted for toxicology testing and assessment encounter a process analogous to that of natural selection, and that this process might be already underway. The authors apply the concept of 'evolutionary pressure' to help us better understand the requirements of future toxicity testing. They acknowledge that current regulatory restrictions were created with animal data in mind, where such data fit in perfectly, leaving little room for new approach methodologies (NAMs) to survive and thrive in a regulatory environment that is 'hostile' to them. However, extinction events occur and that will also be the case for the methods used for toxicity assessments: some toxicity assessment methods will be relevant enough to persist, while others will stop being used and disappear. The regulatory landscape is not static and can be managed and changed (i.e. it is akin to a garden, rather than a natural habitat). Thus, it could potentially be managed to make it a better fit for NAMs. The authors suggest that academia, industry, non-governmental organisations (NGOs), regulators and legislators could act collaboratively as 'gardeners' and shape a regulatory landscape in which NAMs can thrive.Science relies on incremental progress and change happens all the time. In this context, human tissue-based methods appear to be adapting and surviving rather well. Many researchers are keen to make the most of human tissues directly or through data sets stored in online repositoriesfor example, human cells can be used to create reconstituted human airway tissues, which represent important experimental tools to assess functional endpoints in respiratory toxicology. Behrsing and colleagues 2 discussed the need for method standardisation in respiratory toxicology at a workshop, which led to a multi-laboratory