1. One of the main challenges faced by ecologists today is to understand and predict how species interactions will respond to the current environmental change. It is likely that these changes will have a stronger effect on phylogenetic lineages that depend on intimate and specialised ecological interactions, such as most herbivorous insects.

2. In this review, we highlight the aspects that we consider are fundamental for understanding how species interactions change over time. We start by reviewing terminology and conclude that commonly used terms have undesired connotations with regard to what we know about how hosts are acquired and lost over time.

3. Based on that, we suggest host repertoire as a better term to describe the use of multiple hosts than the host range or diet breadth, both of which fail to capture that host use is often non-contiguous and tend to emphasise the total number of hosts, while ignoring host identity.

4. Another important issue that we highlight is the differentiation between fundamental and realised host repertoires, where the latter is the set of hosts that are actually used in nature, whereas the first includes all hosts that can potentially be used by the species. We show that the characterisation of the fundamental host repertoire is key to understanding how insect-plant interactions will respond to the ecological opportunities instigated by environmental change.